JP2003346160A - Device and method for detecting mobile object - Google Patents

Abstract

(57) [Problem] To provide a moving object detection device capable of stably detecting a moving object regardless of noise of a moving image. SOLUTION: A moving image is divided into a plurality of blocks for each frame, and a sum of absolute difference values of pixel values indicating a degree of matching for each block between frames is calculated. A matching degree calculating unit), a moving vector detecting unit 104 that compares the sum of absolute differences for each block of the moving image to detect a moving vector, and a sum of absolute differences corresponding to the moving vector for each block of the moving image. Is compared with a threshold value to switch whether or not to output a moving vector. A moving object detecting unit 109 that detects a moving object with the moving vector output via the moving vector switching unit 108. And a noise measuring unit 105 that measures noise of a moving image, and a threshold value determining unit 107 that determines a threshold value according to the noise measurement result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving object detecting apparatus and a moving object detecting method for detecting a moving object by comparing moving images between frames, and more particularly, to a moving object stably irrespective of noise of the moving image. The present invention relates to a moving object detection device and a moving object detection method capable of detecting a moving object.

[0002]

2. Description of the Related Art A conventional moving object detecting apparatus for detecting a moving object by comparing moving images between frames is disclosed in, for example, JP-A-7-99657. This conventional moving object detection device is designed to suppress erroneous detection of a motion vector in a background region without deteriorating the performance of detecting a motion vector in a moving object region, and different thresholds are set between the moving object region and the background region. The value is used to switch whether or not to output a motion vector.

More specifically, first, a moving image is divided into a plurality of blocks for each frame, and a sum of absolute differences of pixel values indicating a degree of matching for each block between frames is calculated. Next, a motion vector is detected by finding the minimum value of the sum of absolute differences for each block. Next, for each block, a difference between the minimum value, which is the sum of absolute differences, and the sum of absolute differences, which is a simple difference between frames of the same position block, is obtained, and this difference is compared with a threshold value.
Here, if it is equal to or more than the threshold value, a motion vector is output, and if it is less than the threshold value, a zero vector is output. Also, based on the result of the area determination performed based on the motion vector,
As shown in FIG. 16, in the background region 1611, a predetermined first threshold th1 for the background region is used,
In the moving object area 1612, a predetermined second threshold value th2 for the moving object area is used.

[0004]

However, in such a conventional moving object detection apparatus, the threshold value is not an appropriate value depending on the degree of noise of the moving image or the fluctuation of the noise, and the moving vector cannot be obtained. There is a problem that normal output cannot be performed, and therefore, a moving object cannot be detected stably.

The present invention has been made to solve such a problem, and provides a moving object detecting apparatus and a moving object detecting method capable of stably detecting a moving object irrespective of noise of a moving image. Things.

[0006]

A moving object detecting apparatus according to the present invention comprises: a moving image capturing section for capturing a moving image for each frame;
Dividing the moving image into a plurality of blocks for each frame,
A matching degree calculating unit that calculates a matching degree indicating a degree of matching for each block between frames, and a moving vector detecting unit that detects a moving vector by comparing the matching degree for each block of the moving image, For each block of the moving image, a first block of a block corresponding to a start point of the moving vector and a block corresponding to an end point of the moving vector.
A moving vector switching unit that compares a difference between the first matching degree and the second matching degree between the same position blocks with the first matching degree with a threshold value and switches whether or not to output the moving vector; A moving object detecting unit that detects a moving object based on the moving vector output through the moving vector switching unit, a noise measuring unit that measures noise of the moving image, and the threshold according to a result of the noise measurement. And a threshold value determining unit for determining a value.

With this configuration, even if the noise level of the moving image increases, the output of an erroneous moving vector is suppressed, and therefore, the moving image can be stably moved regardless of the degree of the noise of the moving image or the fluctuation of the noise. Objects can be detected.

The moving object detecting apparatus according to the present invention has a configuration in which the noise measuring unit measures noise of the moving image based on a state of distribution of the detected position of the moving vector.

With this configuration, even if the noise level of a moving image rises, the noise level can be measured easily and accurately by paying attention to the distribution state of the detection position of the moving vector. The value can be controlled dynamically and precisely.

In the moving object detecting apparatus according to the present invention, the noise measuring section extracts an isolated block in which no moving vector is detected in an adjacent block from the blocks in which the moving vector is detected, and extracts the moving image. The number of isolated blocks is calculated for each area by dividing the area into a plurality of areas, and the threshold value determining unit determines the threshold value based on the number of isolated blocks for each area.

With this configuration, even if the noise level rises, the noise level can be easily and accurately measured by a simple process of calculating the number of isolated vectors for each region. Without
The threshold can be dynamically and accurately controlled.

The moving object detecting apparatus according to the present invention has a configuration in which the noise measuring section measures noise of the moving image based on a distribution state of at least one of a direction and a magnitude of the moving vector. ing.

With this configuration, even if the noise level of the moving image increases, the noise level can be measured easily and accurately by focusing on the state of the direction and magnitude distribution of the moving vector. The threshold can be dynamically and accurately controlled.

In the moving object detecting apparatus according to the present invention, the noise measuring unit generates frequency distribution information for the direction of the motion vector, determines uniformity of distribution in the direction of the motion vector, The determining unit determines the threshold based on the uniformity.

With this configuration, even if the noise level rises, the noise level can be easily and accurately measured by a simple process of determining the uniformity of the distribution in the direction of the motion vector. , And the threshold value can be dynamically and accurately controlled.

In the moving object detecting apparatus according to the present invention, the noise measuring section generates frequency distribution information for the magnitude of the motion vector, and determines the uniformity of distribution for the magnitude of the motion vector. The threshold value determination unit has a configuration for determining the threshold value based on the uniformity.

With this configuration, even if the noise level increases, the noise level can be easily and accurately measured by a simple process of determining the uniformity of the distribution of the magnitude of the motion vector. The threshold value can be dynamically and precisely controlled without a large load.

The moving object detecting apparatus according to the present invention includes a moving vector information recording section for recording moving vector information on the moving vector detected by the moving vector detecting section, and the noise measuring section stores the moving vector information. For reference, a configuration is provided in which the noise of the moving image is measured based on the state of change of the moving vector.

With this configuration, even if the noise level of the moving image increases, the noise level can be measured easily and accurately by focusing on the state of the change of the motion vector. And it can be controlled accurately.

The moving object detecting apparatus according to the present invention divides the moving image into a plurality of regions, calculates an increase number of the motion vector for each region, and sets the threshold value determination unit to the plurality of increase numbers. In which the threshold value is determined based on the threshold value.

With this configuration, even if the noise level rises, the noise level can be easily and accurately measured by a simple process of calculating the number of increase of the motion vector for each region. The threshold value can be dynamically and accurately controlled without using it.

In the moving object detecting apparatus according to the present invention, the moving object detecting section may add a label to a connected block in which a moving vector is detected in an adjacent block from among the blocks in which the moving vector has been detected. Is detected.

With this configuration, even if the noise level of the moving image increases, the erroneously output moving vector is ignored, and therefore, the motion vector becomes more stable irrespective of the degree of the noise of the moving image and the fluctuation of the noise. Moving object can be detected.

The moving object detection device of the present invention has a configuration in which the moving object detection unit detects a moving object by attaching the same label to a block in which the direction of the motion vector is within a predetermined range. .

With this configuration, the erroneously output motion vector is ignored, and the moving objects moving in different directions are distinguished from each other. Therefore, the moving object can be detected stably. Even when moving objects moving in different directions are mixed, the moving object can be stably detected.

In the moving object detecting apparatus according to the present invention, the moving object detecting section may attach the same label to a block in which the direction of the motion vector is within a certain range and the size of the motion vector is within a certain range. It has a configuration for detecting a moving object.

With this configuration, the erroneously output motion vector is ignored, and the moving objects moving in different directions and at different speeds are distinguished from each other. Therefore, it is possible to stably detect the moving object. It is possible to stably detect a moving object even when moving objects moving in different directions and at different speeds are mixed.

In the moving object detecting apparatus according to the present invention, the matching degree calculating section calculates a sum of absolute differences between pixel values of each block between frames and each block, and the moving vector detecting section calculates the sum of the moving images. A motion vector is detected by obtaining the minimum value of the sum of absolute differences for each block, and the motion vector switching unit determines a first value of a block corresponding to a start point of the motion vector and a block corresponding to an end point of the motion vector. Or the difference between the first sum of absolute differences or the second sum of absolute differences between the same-position blocks is compared with the threshold value.

With this configuration, even when the degree of matching of each block between frames and each block between frames is detected by the sum of absolute differences of pixel values, the movement is stable irrespective of the degree of noise of the moving image or the fluctuation of the noise. Objects can be detected.

[0030]

Embodiments of the present invention will be described below with reference to the drawings.

(First Embodiment) FIG. 1 shows a first embodiment of the present invention.
FIG. 1 is a block diagram illustrating a moving object detection device 100 according to one embodiment. In FIG. 1, a moving image capturing unit 101 captures a moving image output from an imaging device 10 such as a visible camera or a far-infrared camera for each frame. The moving image recording unit 102 records a moving image for each frame. The difference absolute value sum calculation unit 103 (matching degree calculation unit)
The moving image is divided into a plurality of blocks for each frame, and a difference absolute value sum of pixel values indicating a degree of matching is calculated for each block between frames. The motion vector detection unit 104 detects a motion vector by obtaining the minimum value of the sum of absolute differences for each block of the moving image. The noise measurement unit 105 includes the motion vector detection unit 1
The noise of the moving image is measured based on the information on the moving vector detected by the step 04. For example, as information on a motion vector, a detection position of the motion vector,
The noise of a moving image is measured based on any one of the direction of the moving vector and the magnitude of the moving vector, or a combination thereof. The threshold storage unit 106
A plurality of thresholds are stored. The threshold value determination unit 107 determines a threshold value for switching whether or not to output a motion vector, according to the measurement result of the noise. For example, first, a predetermined reference value stored in the threshold storage unit 106 is used as a threshold, and when the noise increases, the threshold switching candidates stored in the threshold storage 106 are selected. , The one corresponding to the state of the noise is selected and used. The motion vector switching unit 108 calculates, for each block of the motion image, the sum of absolute differences between the block corresponding to the start point of the motion vector and the block corresponding to the end point of the motion vector (that is, the sum of absolute differences corresponding to the motion vector). The difference with the sum of absolute differences, which is a simple difference between frames at the same position, is compared with a threshold value. The moving object detection unit 109 detects a moving object based on the motion vector output via the motion vector switching unit 108. Specifically, a moving object is detected by attaching a label to a connected block in which a motion vector has been detected in an adjacent block from among the blocks in which the motion vector has been detected.

The pixel value is a luminance value or other pixel level.

The sum of absolute differences is compared with each pixel value in the block of interest for the captured current frame (hereinafter referred to as “current frame”) and the frame before the current frame (hereinafter referred to as “previous frame”). This is the sum of the absolute values of the differences from the pixel values in the block. This sum of absolute differences is one of the evaluation values (also referred to as “correlation values”) indicating the degree of image matching between the target block and the comparison block (hereinafter, referred to as “matching degree”). The degree of consistency between them is high.

The motion vector indicates the amount of movement of the object. In this embodiment, the comparison block is moved up, down, left, and right on the current frame, and the positional relationship between the comparison block and the block of interest, in which the sum of absolute differences from the block of interest in the previous frame is minimum, is set as a motion vector. It has become.

The noise is generated in an image pickup apparatus, a transmission path of a moving image, or the like, and is a change in a pixel value independent of a change in an object to be imaged.

Next, the operation of the moving object detecting apparatus according to this embodiment will be described with reference to the flowchart of FIG.

First, a moving image is captured for each frame (S
201). The captured moving image is stored in the moving image recording unit 102.
Is recorded for each frame.

Next, the moving image is divided into a plurality of blocks for each frame, and the difference absolute value of the pixel value between each block of the previous frame extracted from the moving image recording unit 102 and each block of the current frame captured is extracted. Calculate the sum (S2
02). Specifically, in the vicinity of each block of interest,
By moving the comparison block on the current frame from -X to X in the horizontal direction and from -Y to Y in the vertical direction,
As shown in, sum of absolute difference V (v _x, v _y) for each pixel value I _t of each comparison in a block of pixel values I _t-1 and the current frame in each target block in the previous frame, respectively calculate.

(Equation 1) Here, (b _x , b _y ) is the position of each block with respect to the reference block in the frame, (m, n) is the size of each block, and (x, y) is the size of each block. Is the relative position of each pixel with each reference point of the origin as (v _x ,
v _y ) is a vector from each block of interest to each contrast block.

Next, for each block of the moving image, a moving vector is detected by obtaining the minimum value of the sum of absolute differences (S2).
03). Specifically, as shown in Expression 2, the minimum value V _{min of the} sum of absolute differences between each block of interest and each of the comparison blocks is determined, and a vector (v _x , V _y ) as a motion vector.

(Equation 2) Next, the noise of the moving image is measured based on the information on the detected moving vector (S204). As for how to measure the noise, (1) the noise of the moving image is measured based on the state of the distribution at the detection position of the moving vector, and (2) the distribution of the direction and the size of the moving vector is measured. Measure the noise of the video based on the state,
(3) Noise of a moving image is measured based on a state of change of a moving vector in a time series. Specific examples of these noise measurements will be described in detail in each of the second to fourth embodiments described later.

Next, a threshold value for switching whether or not to output a motion vector is determined according to the noise measurement result (S205). Specifically, the threshold storage unit 106
Is used as an initial threshold value, and when noise increases, the threshold value storage unit 1
From the threshold switching candidates stored in step 06, those corresponding to the noise state are selected and used.

Next, for each block of the moving picture, difference absolute value sum V corresponding to the inter-frame difference between blocks of the differential absolute value sum V _{mi n} at the same position corresponding to the motion vector
_By comparing the difference V _(0,0) −V _min with _(0,0) with a threshold value,
Switching whether to output a motion vector or not (S20)
6). FIG. 11 shows the sum of absolute differences V (v _x , v _y ) between each block of interest in the previous frame and each comparison block in the current frame.
Original motion vector information 1110 obtained by finding the minimum value of
13 shows an example of moving vector information 1120 for detecting a moving object in which whether or not to output is switched by comparison with a threshold. 11, in the original vector information 1110, a plurality of motion vectors due to movement of an object are detected in the moving object region 1112, and a plurality of motion vectors due to noise are detected in the background region 1111. In the present embodiment, the threshold is determined according to the state of the noise, and the difference between the sum of absolute differences corresponding to each motion vector and the sum of absolute differences between blocks at the same position is equal to or greater than the threshold. By switching so as to output a motion vector at a certain time and not to output a motion vector when the value is smaller than the threshold value (output a zero vector), the motion vector information 1120 for detecting a moving object is A plurality of motion vectors due to the movement of the object are output, and almost no motion vector due to noise is output in the background area 1121.

Next, a moving object is detected from the output motion vector (S207). Specifically, a moving object is detected by attaching a label to a connected block in which a motion vector has been detected in an adjacent block from among the blocks in which the motion vector has been detected.

The above steps are continued until the moving image ends (S208).

The steps shown in FIG. 2 may be executed by a computer.

The step of measuring noise (S20)
4) and the example of performing the step (S206) of switching the motion vector after the step of determining the threshold value (S205) and the step of detecting the moving object (S207), the steps of switching the motion vector (S206) and Step of detecting a moving object (S2
07), a step of measuring noise (S204) and a step of determining a threshold (S205) are performed.
The motion vector may be switched at the newly determined threshold from the next frame.

In the step (S206) for switching the motion vector, the sum of absolute differences V _min between blocks at the same position as the sum of absolute differences V _min corresponding to the motion vector is obtained.
_{The example in} which the difference V _(0,0) −V _min from _(0,0) is compared with the threshold value has been described, but the sum of the absolute difference value V _min corresponding to the motion vector is compared with the threshold value. You may.

Also, an example in which the threshold value is dynamically changed by extracting the threshold value candidate from the threshold value storage unit 106 has been described. May be determined in advance, and the threshold value may be dynamically changed. In addition, this includes the case where there are two types of thresholds.

As described above, the moving object detecting apparatus according to the present embodiment measures the noise of a moving image and sets the threshold for switching whether or not to output a moving vector according to the result of the noise measurement. Therefore, even if the noise level of the moving image increases, the output of an erroneous moving vector is suppressed, and therefore, the moving object can be stably detected regardless of the degree of the noise of the moving image or the fluctuation of the noise. Can be detected.

Since the noise is measured based on the information on the motion vector, the noise affecting the detection of the motion vector can be easily and accurately measured. Therefore, the threshold value can be dynamically and accurately controlled. can do.

Further, since a moving object is detected by attaching a label to a connected block in which a motion vector is detected in an adjacent block from among blocks in which a motion vector is detected, a motion vector of an erroneously output isolated block is obtained. Is ignored, and therefore, the moving object can be detected more stably.

(Second Embodiment) FIG. 3 is a block diagram showing a moving object detection apparatus according to a second embodiment of the present invention. In FIG. 3, the same components as those of the moving object detection device according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and these components are described in the first embodiment. Therefore, detailed description is omitted.

In the present embodiment, the position distribution measuring section 3
05 measures the noise of the moving image based on the state of distribution of the detection position of the moving vector.
Specifically, an isolated block in which a motion vector is not detected in an adjacent block is extracted from blocks in which a motion vector is detected, and a moving image is divided into a plurality of regions, and the number of isolated blocks is calculated for each region. I do. The threshold value determination unit 307 determines a threshold value based on the number of isolated blocks for each area.

Next, the operation of the moving object detecting apparatus according to the present embodiment will be described with reference to the flowchart of FIG.

In FIG. 4, the steps from the step of taking in a moving image (S201) to the step of obtaining a minimum value of the sum of absolute differences for each block of the moving image to detect a moving vector (S203) are the first embodiment. This is the same processing as each step described in the embodiment.

Next, an isolated block in which no motion vector is detected is extracted from adjacent blocks in the vicinity of four or eight neighboring blocks from among the blocks in which the motion vector has been detected, and the moving image is equally divided into a plurality of regions. Then, the number of isolated blocks is calculated for each area (S404). Next, a threshold for switching whether or not to output a motion vector is determined according to the number of isolated blocks for each area (S405).

For example, based on the number of isolated blocks for each area, it is determined whether or not the isolated blocks are evenly distributed in the entire area. When the blocks are evenly distributed, the threshold value is determined from a predetermined reference value. Change the threshold to a value greater than the value. FIG. 12 shows an example in which isolated blocks are evenly distributed in the entire area. In FIG. 12, the number of isolated blocks in each area is equal to or larger than a predetermined first number (for example, “3”).
And the difference between the maximum value of the number of isolated blocks in each area (“4” in FIG. 12) and the minimum value of the number of isolated blocks in each area (“3” in FIG. 12). In 12, it is determined whether or not “1” is equal to or less than a predetermined second number (for example, “4” or less), and the number of isolated blocks in each area is equal to or greater than the first number. If the difference between the maximum value of the number of isolated blocks and the minimum value of the number of isolated blocks is equal to or smaller than the second number, the threshold is changed to a large threshold.

In FIG. 4, a step of switching whether or not to output a motion vector (S206), a step of detecting a moving object with a motion vector (S207), and a step of determining whether or not to end (S208) ) Means
This is a process similar to each step described in the first embodiment.

The steps shown in FIG. 4 may be executed by a computer.

In the case where the noise is reduced and the number of isolated blocks is reduced, for example, when there is an area where a motion vector is not detected, the threshold value may be returned to the reference value.

As described above, even if the noise level of the moving image increases, the moving object detection apparatus of the present embodiment can easily reduce the noise level by paying attention to the distribution state of the detection position of the motion vector. In this case, the threshold value can be measured accurately, so that the threshold value can be dynamically and precisely controlled. Further, by calculating the number of isolated vectors for each region, the noise level can be easily and accurately measured, and therefore, it is possible to dynamically and accurately control the threshold value without much load. it can.

(Third Embodiment) FIG. 5 is a block diagram showing a moving object detection apparatus according to a third embodiment of the present invention. In FIG. 5, the same components as those of the moving object detection device according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and these components are described in the first embodiment. Therefore, detailed description is omitted.

In this embodiment, the frequency distribution measuring section 5
05 generates information indicating the distribution of the direction of the motion vector and the magnitude of the motion vector, and determines the uniformity of the distribution for at least one of the distribution of the direction of the motion vector and the magnitude of the motion vector. ing. The threshold value determination unit 507 determines a threshold value based on the uniformity obtained by the frequency distribution measurement unit 505.

Next, the operation of the moving object detecting apparatus according to the present embodiment will be described with reference to the flowchart of FIG.

In FIG. 6, the steps from the step of capturing a moving image (S201) to the step of obtaining a minimum value of the sum of absolute differences for each block of the moving image to detect a moving vector (S203) are the first embodiment. This is the same processing as each step described in the embodiment.

Next, a histogram (frequency distribution information) for the direction of the motion vector is generated, the uniformity of distribution is determined for the direction of the motion vector (S604), and a threshold value is determined based on the uniformity. (S605).

For example, a moving image is equally divided into a plurality of regions, and in all the regions, a moving vector is detected in blocks having a fixed ratio or more with respect to the total number of blocks in each region. When it is determined that each frequency in the direction is within a certain range, the threshold is changed to a larger threshold. FIG. 13 shows motion vector information 1310 which is a source when a motion vector due to noise is detected in all regions when there is no moving object. As described above, when the motion vector due to noise has increased in the entire region as compared with the case where there is no influence due to noise or the case where the influence is small, the noise becomes large based on the histogram 1401 as shown in FIG. Judgment has been made, and the threshold is changed. In the histogram 1401 of FIG. 14, it is determined that each frequency in the direction of the motion vector is in the range from “5” to “7” and is distributed substantially uniformly, and the threshold is changed to a large threshold. Is done.

In FIG. 6, a step of switching whether or not to output a motion vector (S206), a step of detecting a moving object with a motion vector (S207), and a step of determining whether or not to end (S208) ) Means
This is a process similar to each step described in the first embodiment.

The steps shown in FIG. 6 may be executed by a computer.

When the noise is reduced and the direction and size are varied, for example,
When there is an area where no motion vector is detected, the threshold value may be returned to the reference value.

As described above, the moving object detecting apparatus according to the present embodiment can reduce the noise level by paying attention to the direction and size distribution of the motion vector even if the noise level of the moving image increases. It will be easy and accurate to measure and therefore the threshold can be dynamically and accurately controlled. In addition, by determining the uniformity of the distribution of the direction of the motion vector and the magnitude of the motion vector, the noise level can be easily and accurately measured, and therefore, the threshold value can be set without much load. It can be controlled dynamically and accurately.

Although the example in which the threshold value is determined based on the state of distribution of the direction of the motion vector has been described, the threshold value is determined based on the state of distribution of the direction of the motion vector and the size of the motion vector. May be determined. Further, it includes a method in which the noise of a moving image is measured based on the state of distribution of the magnitude of the moving vector, and the noise of the distribution in the direction of the moving vector is not measured.

(Fourth Embodiment) FIG. 7 is a block diagram showing a moving object detection apparatus according to a fourth embodiment of the present invention. 7, the same components as those of the moving object detection device according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and these components will be described in the first embodiment. Therefore, detailed description is omitted.

In the present embodiment, the motion vector information recording unit 710 records motion vector information on the motion vector detected by the motion vector detection unit 104. The motion vector information includes at least a detection position of the motion vector. The motion vector information may include the direction and magnitude of the motion vector. Change measurement unit 7
In reference numeral 05, the noise of the moving image is measured based on the state of the change of the moving vector with reference to the moving vector information. Specifically, the moving image is divided into a plurality of regions, and the number of increase of the moving vector is calculated for each region. The threshold value determining unit 707 determines a threshold value based on the number of increase of the motion vector for each region.

Next, the operation of the moving object detecting apparatus according to the present embodiment will be described with reference to the flowchart of FIG.

In FIG. 8, the steps from the step of taking in a moving image (S201) to the step of finding a moving vector by obtaining the minimum value of the sum of absolute differences for each block of the moving image (S203) are the first embodiment. This is the same processing as each step described in the embodiment.

Next, the moving image is divided into a plurality of regions, and the increment number of the moving vector is calculated for each region (S804). Next, the threshold value is determined based on the number of increase of the motion vector for each region (S805).

For example, when the number of motion vectors increases in all regions, the threshold value is changed from a predetermined reference value to a threshold value larger than this reference value.

The steps shown in FIG. 8 may be executed by a computer.

When the noise is reduced and the number of motion vectors is reduced, for example, when there is an area where no motion vector is detected, the threshold value may be returned to the reference value.

In FIG. 8, a step of switching whether or not to output a motion vector (S206), a step of detecting a moving object with the motion vector (S207), and a step of determining whether or not to end (S208) ) Means
This is a process similar to each step described in the first embodiment.

As described above, the moving object detecting apparatus according to the present embodiment can easily and accurately measure the noise level by paying attention to the state of the change of the motion vector even if the noise level of the moving image increases. And therefore,
The threshold value can be controlled dynamically and accurately.

Further, the noise level can be easily and accurately measured by calculating the number of increase of the motion vector for each region. Therefore, the threshold value can be dynamically and accurately determined without much load. Can be controlled.

(Fifth Embodiment) FIG. 9 is a block diagram showing a moving object detection apparatus according to a fifth embodiment of the present invention. In FIG. 9, the same components as those of the moving object detection device according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and these components will be described in the first embodiment. Therefore, detailed description is omitted.

In the present embodiment, the moving object detector 9
09 is a connected block in which the motion vector is detected in an adjacent block from among the blocks in which the motion vector is detected, and the same label is assigned to the block in which the direction of the motion vector is within a certain range. A moving object is detected.

Next, the operation of the moving object detecting apparatus according to the present embodiment will be described with reference to the flowchart of FIG.

In FIG. 10, the steps from the step of capturing a moving image (S201) to the step of switching whether or not to output a moving vector (S206) are the same as the steps described in the first embodiment. .

In the present embodiment, four adjacent blocks or eight adjacent blocks are selected from blocks in which a motion vector is detected.
A moving object is detected by attaching the same label to a connected block in which a motion vector is detected in a nearby block and the direction of the motion vector is within a predetermined range (S
1007). FIG. 15 shows an example of labeling. A first moving object area 1521 in which each block is labeled based on the motion vector information 1510 indicating a detection position, a direction, and a magnitude of the motion vector, and a label 'a' is added;
The second moving object area 1522 labeled “b” is detected.

The above steps are continued until the moving image ends (S208).

The steps shown in FIG. 10 may be executed by a computer.

As described above, in the moving object detection apparatus according to the present embodiment, even if the noise level of a moving image increases, the erroneously output moving vector is ignored, and therefore, the moving image The moving object can be detected more stably irrespective of the degree of the noise or the fluctuation of the noise.

Also, since a moving object is detected by attaching the same label to a block in which the direction of the motion vector is within a certain range, moving objects moving in different directions are distinguished from each other, and moving in different directions. It is possible to stably detect a moving object even when there are moving objects mixed.

In the fifth embodiment, an example has been described in which a moving object is detected by assigning the same label to a block in which the direction of a motion vector is within a certain range. However, the present invention is not limited to this. , A block in which the direction of the motion vector is within a certain range and the magnitude of the motion vector is within a certain range is assigned the same label to detect a moving object.

As described above, when a moving object is detected by attaching the same label to a block whose magnitude of the motion vector is within a certain range, the moving objects moving at different speeds are distinguished from each other. Thus, even when moving objects that move at different speeds are mixed, the moving object can be stably detected.

In the first to fifth embodiments, examples have been described in which the degree of matching between blocks is calculated as the sum of absolute differences between pixel values. However, the present invention is not limited to this. Instead, it includes the calculation of the degree of matching between blocks based on normalized cross-correlation values and other evaluation values of matching.

[0095]

According to the present invention, it is possible to provide a moving object detection apparatus and a moving object display method which have an excellent effect that a moving object can be detected stably irrespective of noise of a moving image. .

[Brief description of the drawings]

FIG. 1 is a block diagram showing a moving object detection device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing moving object detection processing in the moving object detection device according to the first embodiment of the present invention;

FIG. 3 is a block diagram showing a moving object detection device according to a second embodiment of the present invention;

FIG. 4 is a flowchart showing moving object detection processing in the moving object detection device according to the second embodiment of the present invention;

FIG. 5 is a block diagram showing a moving object detection device according to a third embodiment of the present invention.

FIG. 6 is a flowchart showing a moving object detection process in the moving object detection device according to the third embodiment of the present invention.

FIG. 7 is a block diagram showing a moving object detection device according to a fourth embodiment of the present invention.

FIG. 8 is a flowchart showing a moving object detection process in the moving object detection device according to the fourth embodiment of the present invention.

FIG. 9 is a block diagram showing a moving object detection device according to a fifth embodiment of the present invention.

FIG. 10 is a flowchart illustrating moving object detection processing in the moving object detection device according to the fifth embodiment of the present invention.

FIG. 11 is an explanatory diagram illustrating determination of a threshold value in the moving object detection device according to the first embodiment of this invention.

FIG. 12 is an explanatory diagram illustrating a state of distribution of detection positions of motion vectors in the moving object detection device according to the second embodiment of this invention.

FIG. 13 is an explanatory diagram for explaining a state of distribution of directions of motion vectors in the moving object detection device according to the third embodiment of the present invention.

FIG. 14 is a diagram showing frequency distribution information in the direction of a motion vector in the moving object detection device according to the third embodiment of the present invention.

FIG. 15 is an explanatory diagram for describing moving object detection in the moving object detection devices according to the first to fifth embodiments of the present invention.

FIG. 16 is an explanatory diagram for explaining a threshold value in a conventional moving object detection device.

[Explanation of symbols]

101 Moving Image Capture Unit 102 Moving image recording unit 103 Difference absolute value sum calculation unit (matching degree calculation unit) 104 Motion vector detector 105 Noise measurement unit 106 threshold storage 107 Threshold value determination unit 108 Motion vector switching unit 109 Moving object detector

Claims (21)

[Claims] 1. A moving image capturing unit that captures a moving image for each frame, a matching degree indicating a degree of matching for each block between frames, dividing the moving image into a plurality of blocks for each frame. A matching degree calculating unit for calculating
A moving vector detecting unit that compares the degree of matching for each block of the moving image to detect a moving vector, and for each block of the moving image, a block corresponding to a starting point of the moving vector and an ending point of the moving vector. The first matching degree with the block or the difference between the first matching degree and the second matching degree between the same position blocks are compared with a threshold value to switch whether to output the motion vector. A motion vector switching unit, a moving object detection unit that detects a moving object with the motion vector output via the motion vector switching unit, and a noise measurement unit that measures noise of the moving image,
A moving object detecting device, comprising: a threshold value determining unit that determines the threshold value according to the measurement result of the noise. 2. The noise measuring unit according to claim 1, wherein the noise measuring unit measures the noise of the moving image based on a distribution state of a detection position of the moving vector.
3. The moving object detection device according to claim 1. 3. The noise measuring unit extracts an isolated block in which a motion vector is not detected in an adjacent block from blocks in which the motion vector is detected,
The moving image is divided into a plurality of regions, the number of isolated blocks is calculated for each region, and the threshold value determination unit determines the threshold value based on the number of isolated blocks for each region. The moving object detection device according to claim 2, wherein: 4. The noise measuring unit according to claim 1, wherein the noise measuring unit measures the noise of the moving image based on a distribution state of at least one of a direction and a magnitude of the moving vector. 3. The moving object detection device according to claim 1. 5. The noise measuring unit generates frequency distribution information for the direction of the motion vector, determines uniformity of distribution in the direction of the motion vector, and the threshold value determining unit determines the uniformity of the distribution. The moving object detection device according to claim 4, wherein the threshold value is determined based on the following. 6. The noise measurement unit generates frequency distribution information for the size of the motion vector, determines the uniformity of distribution for the size of the motion vector, and the threshold value determination unit 6. The moving object detection device according to claim 4, wherein the threshold value is determined based on uniformity. 7. A motion vector information recording unit for recording motion vector information on the motion vector detected by the motion vector detection unit, wherein the noise measuring unit refers to the motion vector information, The moving object detection device according to claim 1, wherein the noise of the moving image is measured based on a state of change of the moving image. 8. The noise measurement unit divides the moving image into a plurality of regions, calculates an increase number of the motion vector for each region, and the threshold value determination unit calculates the increase number of the motion vector The moving object detection device according to claim 7, wherein the threshold value is determined based on the threshold value. 9. The moving object detection unit detects a moving object by attaching a label to a connected block in which a motion vector is detected in an adjacent block from among the blocks in which the motion vector is detected. The moving object detection device according to claim 1, wherein 10. The moving object detection unit according to claim 9, wherein the moving object detection unit detects the moving object by attaching the same label to a block in which the direction of the motion vector is within a predetermined range. Moving object detection device. 11. The moving object detection unit detects a moving object by attaching the same label to a block in which the direction of the motion vector is within a certain range and the magnitude of the motion vector is within a certain range. The moving object detection device according to claim 9, wherein 12. The matching degree calculating unit calculates a difference absolute value sum of pixel values for each block between frames and each block, and the moving vector detecting unit calculates the difference absolute value for each block of the moving image. The motion vector is detected by obtaining the minimum value of the sum, and the motion vector switching unit is configured to calculate a first sum of absolute differences between the block corresponding to the start point of the motion vector and the block corresponding to the end point of the motion vector, or 12. The method according to claim 1, wherein a difference between a first sum of absolute difference values and a second sum of absolute difference values between blocks at the same position is compared with the threshold value. The moving object detection device according to claim 1. 13. A moving image capturing step of capturing a moving image for each frame, and a matching degree indicating a degree of matching for each block between frames by dividing the moving image into a plurality of blocks for each frame. Calculating a matching level, calculating a matching vector for each block of the moving image to detect a moving vector, a noise measuring step of measuring noise of the moving image, And a threshold value determining step of determining a threshold value for switching whether or not to output based on the measurement result of the noise, and for each block of the moving image, a block corresponding to a starting point of the moving vector and the A first degree of matching with a block corresponding to the end point of the motion vector or a difference between the first degree of matching and a second degree of matching between blocks at the same position Moving object switching step of comparing whether or not to output the moving vector by comparing the minute with a threshold value, and a moving object detecting step of detecting a moving object with the moving vector. Detection method. 14. A moving image capturing step of capturing a moving image for each frame, and a matching degree indicating a degree of matching for each block between frames by dividing the moving image into a plurality of blocks for each frame. Calculating a matching degree, calculating a matching degree for each block of the moving image to detect a moving vector, and selecting a moving vector from among the blocks in which the moving vector is detected to an adjacent block. Extracting an isolated block where no is detected, dividing the moving image into a plurality of regions, and calculating the number of the isolated blocks for each region, based on the number of isolated blocks for each region. A threshold value determining step of determining a threshold value for switching whether or not to output the moving vector; and For a block, a first matching degree between a block corresponding to a start point of the motion vector and a block corresponding to an ending point of the motion vector, or a difference between the first matching degree and a second matching degree between blocks at the same position. A moving object switching step of comparing whether or not to output the moving vector, and a moving object detecting step of detecting a moving object with the moving vector. Method. 15. A moving image capturing step of capturing a moving image for each frame, and a matching degree indicating a degree of matching for each block between frames by dividing the moving image into a plurality of blocks for each frame. Calculating a degree of matching, calculating a moving vector by comparing the degree of matching for each block of the moving image to detect a moving vector, and frequency distribution information for at least one of a direction and a size of the moving vector. A frequency distribution information generating step of generating, a threshold value determining step of determining a threshold value for switching whether or not to output the motion vector based on the frequency distribution information, and each block of the moving image , There is a first degree of matching between the block corresponding to the start point of the motion vector and the block corresponding to the end point of the motion vector. A moving vector switching step of comparing the difference between the first matching degree and the second matching degree between the same position blocks with a threshold value to switch whether or not to output the moving vector; A moving object detection step of detecting a moving object by a vector. 16. A moving image capturing step of capturing a moving image for each frame, and a matching degree indicating a degree of matching for each block between frames by dividing the moving image into a plurality of blocks for each frame. Calculating a matching degree, calculating a matching degree for each block of the moving image to detect a moving vector, and dividing the moving image into a plurality of regions to determine the moving image for each region. A motion vector increase number calculation step of calculating an increase number of vectors, and a threshold value determination step of determining a threshold value for switching whether to output the motion vector based on the plurality of increase numbers, For each block of the moving image, a first block of a block corresponding to a starting point of the moving vector and a block corresponding to an ending point of the moving vector.
A moving vector switching step of comparing a difference between the first matching degree or the second matching degree between the same position blocks with the first matching degree with a threshold value and switching whether or not to output the moving vector; A moving object detection step of detecting a moving object with the motion vector. 17. The moving object detecting step detects the moving object by attaching a label to a connected block in which a motion vector is detected in an adjacent block from among the blocks in which the motion vector is detected. Claim 1
The moving object detection method according to any one of claims 3 to 16. 18. The moving object according to claim 17, wherein, in the moving object detecting step, the moving object is detected by attaching the same label to a block in which the direction of the motion vector is within a predetermined range. Object detection method. 19. The moving object detecting step includes detecting the moving object by attaching the same label to a block in which the direction of the motion vector is within a certain range and the magnitude of the motion vector is within a certain range. The moving object detection method according to claim 17, wherein: 20. The matching degree calculating step calculates a difference absolute value sum of pixel values for each block between frames and each block, and the moving vector detecting step calculates the difference absolute value for each block of the moving image. A motion vector is detected by calculating a minimum value of the sum, and the motion vector switching step includes a first sum of absolute differences between a block corresponding to a start point of the motion vector and a block corresponding to an end point of the motion vector, or 20. The difference between the first sum of absolute differences and the second sum of absolute differences between same-position blocks is compared with the threshold value.
The moving object detection method according to any one of the above. 21. A moving image capturing step of capturing a moving image for each frame, and a matching degree indicating a degree of matching for each block between frames by dividing the moving image into a plurality of blocks for each frame. Calculating a matching level, calculating a matching vector for each block of the moving image to detect a moving vector, a noise measuring step of measuring noise of the moving image, A threshold value determining step of determining a threshold value for switching whether or not to output according to the measurement result of the noise, and for each block of the moving image, a block corresponding to a starting point of the moving vector and the A first degree of matching with the block corresponding to the end point of the motion vector or a difference between the first degree of matching and the second degree of matching between blocks at the same position Comparing a minute with a threshold value, and causing a computer to execute a moving vector switching step of switching whether or not to output the moving vector, and a moving object detecting step of detecting a moving object with the moving vector. Moving object detection program.