TWI436038B - Light intensity image detection system and its detection method - Google Patents

Description

Light intensity image detection system and detection method thereof

The invention relates to an image detection system, in particular to a light intensity image detection system and a detection method thereof.

At present, the image detection system uses a dual camera to establish a stereo vision system, and uses the concept of parallax between the planes of the two cameras, combined with the extraction of feature points, and then uses the feature points to reconstruct a three-dimensional space to create an obstacle image. For the application of such systems, the most common is laser ranging, laser ranging is a method of energy reflection receiving to detect, the principle of laser ranging is similar to the principle of general ultrasonic reflection: by the transmitter After a pulse wave is first emitted, the reflected wave after hitting the obstacle returns to the receiver according to the original path, and the distance between the obstacle and the laser range finder can be obtained according to the time interval between the transmitted pulse wave and the received echo.

The basic method of laser ranging is to emit a laser pulse to an object to be measured and start timing, and stop timing when receiving the reflected light. This time can be converted to the distance between the laser and the target. That is, a beam is emitted to the target, and the target reflected light wave returns to the detector, and the time taken for the light wave to go back and forth is converted into a distance. The laser range finder can also emit multiple laser pulses, using the Doppler effect to determine if the object is moving away or near the source. However, since the speed of light is too fast, the instrument is not accurate when measuring close distances, and triangulation is more suitable when high-precision ranging of less than a millimeter is required. In addition, in the method of detecting obstacles by lightning, the laser range finder will return the detected distance of the laser light emitted from each angle, and convert the data of each distance into the rectangular coordinates to obtain the data. Obstacle image.

At present, in the development of robots, the technology of navigation systems is mostly developed by laser range finder. When autonomous mobile robots move in an unknown environment, they will hinder their advancement due to surrounding dynamic or static obstacles, and when they collide with obstacles. , can cause damage to robots or environmental items. Therefore, the environmental obstacle sensing technology of smart mobile robots is very important. According to the movement characteristics of the robot with differential motion, the position of the obstacle in front of the robot will be an important factor affecting the control of the forward direction of the robot. The laser range finder, which performs a flat scan with a laser, constructs the obstacle structure and environment around the robot through the distance value of each sampling angle. By acquiring the obstacle structure around the robot, the robot steering control can be performed through algorithms such as fuzzy reaction navigation, so that the robot can avoid obstacles and continue to advance.

In addition, since the laser range finder can detect the environmental topography around the robot, this technology can also be combined with the robot's positioning system to draw an environmental map. However, the laser range finder around the people currently on the market is quite expensive (up to NT$70,000) and must be imported from abroad. As a result, it has also caused an increase in the cost of domestic R&D mobile robots and difficulties in the procurement and maintenance of key components. Finally, the laser rangefinder is an indispensable key ICT component for autonomous mobile robots. However, due to the expensive installation cost, most autonomous mobile robots still use low-cost sensors such as ultrasonic or infrared. It causes restrictions on the robot's autonomous obstacle avoidance and navigation capabilities. Taiwan's optoelectronic industry is booming, and laser technology is one of the key technologies in the optoelectronic industry. Therefore, how to develop a low-cost image/distance instrument based on the existing optoelectronic technology is the current industry demand for improvement.

In view of the above-mentioned background of the invention, in order to meet the needs of the industry's special interests, the present invention provides an image detection system using the intensity of the light source to solve the above-mentioned problems that the conventional art has failed to achieve.

An object of the present invention is to provide a light intensity type image detecting system and a detecting method thereof, which are a range of halos formed by a light source illuminating an obstacle, and the halo change indicates a change intensity on the image. As shown in the first figure, the halo is concentric shape, with the center of the concentric circle being the brightest, and the gradient gradually darkening. The energy of the halo is gradually weakened outward at the center, and this feature can be obtained. Know the existence of obstacles and their distance. According to its halo characteristics, the center of the halo is the strongest part of the overall halo energy. When the obstacle is large in the distant range of the light source, the range of the strongest energy in the center is small; on the contrary, when the obstacle is in the light source In the vicinity, the overall halo range is small, but the center has the strongest energy range. From this central energy range, it is known whether there are obstacles and their proximity, and combined with a single camera to detect the distance of obstacles and obstacles. The present invention utilizes a projection light source such as an LED to detect the presence and distance of an obstacle, and the projection light source forms a halation on the obstacle. As shown in the second figure, in the field of digital image processing, the image presents intensity. The invention uses the energy of the light source to change the intensity in the image, and combines the algorithm of digital image processing to determine the existence and proximity of the obstacle. Accordingly, the present invention can replace the conventional architecture of positioning images with dual cameras and replace expensive laser detection devices. Therefore, if a low-cost image/distance instrument can be developed on the basis of the existing photovoltaic technology in the country, not only can the cost of the autonomous mobile machine be greatly reduced, the navigation stability can be improved, and the product acceptance can be increased. The establishment of a larger product competitive advantage in the domestic optoelectronic industry, according to which the invention is in line with the value of industrial application and innovation.

In accordance with the purpose of the present invention, the present invention provides an optical intensity image detection system that includes an image capture module for performing an image capture step to generate a first image data; an information processing module, the information processing module receives the first image data to perform an intensity calculation step and generates an intensity mean value; and a control module, the control module receives the intensity average value A statistical value is generated, wherein the control module can generate a read signal according to the statistical value.

The image capture module further includes a light source projection device for projecting a light source outside the light intensity image detection system, and the light source projection device further includes a light emitting diode component, and the image is The module further includes an image receiving device for receiving the light source to reflect and generating the first image data. In addition, the information processing module further includes a grayscale unit that receives the first image data to generate a second image data, and a halo quantification unit that receives the second image data. Image data to generate a third image data; and a pixel calculation unit, the pixel calculation unit receives the third image data to generate the intensity mean, wherein the grayscale unit performs a grayscale step to convert The color information of the first image data is the second image data, and the halo quantifying unit performs a halo calculation step to determine a halo range of the image, and simultaneously removes the non-halo area to form the third image data. The pixel calculation unit performs the intensity calculation step to calculate an average of the intensities of all the pixels of the third image data and generate the intensity mean.

The control module further includes an IT calculation unit for receiving the intensity average value and performing an IT statistical step to generate the statistical value, wherein the IT statistics step is an IT calculation unit that records the information processing module according to time. The intensity average output by time. In addition, the control module further includes a control unit, the control unit receives the statistical value to perform a reading step and generates the interpretation signal, wherein the determining step is that the control unit determines whether the image is close to the statistical value according to the statistical value. The light intensity type image detecting system, and the above-mentioned interpretation step determines that the intensity mean becomes stronger with time, the interpretation signal indicates that the image is close to the light intensity type image detecting system; and the reading step determines the intensity mean value If the time is weak, the interpretation signal indicates that the image is away from the light intensity image detection system; and the interpretation step determines that the intensity average does not continuously increase/decrement with time, the interpretation signal indicates the light intensity image. There is no image before the detection system. In addition, the control unit performs a control step according to the interpretation signal to generate a control signal, and the light intensity image detection system further includes an operation module, and the operation module receives the control signal, wherein the operation mode is The group further includes a power unit and a transmission unit, wherein the operation module receives the control signal to operate and control the power unit, thereby driving the transmission unit, and the power unit further includes a motor.

According to an aspect of the present invention, the present invention provides a light intensity image detecting method, which comprises performing an image capturing step to receive a light source reflection and generate a first image data; performing a grayscale Step: converting color information of the first image data to generate a second image data; performing a halo calculation step according to the second image data to determine a halo range of the image, simultaneously removing the non-halo area and generating a third And performing an intensity calculation step according to the third image data to calculate an average of the intensities of all pixels of the third image data and generating an intensity mean, wherein the intensity mean is continuously generated over time. The light intensity image detection method described above is performed by a light intensity image detection system.

The light intensity image detecting method further includes a light source projecting step of projecting a light source by a light source projection device, wherein the image capturing step is performed by an image receiving device to receive a light source reflection and generate The first image data. In addition, the grayscale step is to generate the second image data by converting the color information of the first image data by a grayscale unit, and the halo calculation step is performed by the halo quantification unit. The second image data determines a halo range of the image, and removes the non-halo area to generate the third image data, and the intensity calculation step is performed by the pixel calculation unit to calculate the third image according to the third image data. The mean of the intensity of all pixels of the data and produces the mean of the intensity. Furthermore, the light intensity image detection method further includes an IT statistical step of performing the IT statistical step by an IT calculation unit according to the intensity average outputted by the time record according to the time record, and generating a statistical value, and the The light intensity image detecting method further includes a reading step, wherein the reading step is performed by the control unit to receive the statistical value, and generates a reading signal, wherein the intensity average of the statistical value becomes stronger with time. The interpretation signal indicates that the image is close to each other; if the intensity average of the above statistical value becomes weaker with time, the interpretation signal indicates that the image is far away; the intensity average of the above statistical value does not continuously increase/decrement with time. If the change is made, the interpretation signal indicates no image. In another aspect, the light intensity image detecting method further includes a control step, wherein the control module generates a control signal according to the read signal by a control unit, wherein the control signal is used to control a power unit to drive a control unit. Actuation of the drive unit.

The direction of the invention discussed herein is an image detection system. In order to fully understand the present invention, detailed structures and elements thereof will be set forth in the following description. Obviously, the implementation of the present invention is not limited to the specific details familiar to those skilled in the art of image detection systems. On the other hand, well-known structures and elements thereof are not described in detail to avoid unnecessary limitation of the invention. The preferred embodiments of the present invention are described in detail below, but the present invention may be widely practiced in other embodiments, and the scope of the present invention is not limited by the scope of the following patents. .

Referring to the third and fourth embodiments, the present invention provides a light intensity image detection system 100. The light intensity image detection system 100 includes an image capture module 120. An image processing module 140 and a control module 160, wherein the image capturing module 120 performs an image capturing step 210 to generate a first image data 210A, and the information processing module 140 receives the first image data 210A. A grayscale step 220 is performed to generate a second image data 220A, and a halo calculation step 230 is performed according to the second image data 220A to generate a third image data 230A, and then an intensity calculation is performed according to the third image data 230A. Step 240 is to generate an intensity mean In, and the control module 160 receives the intensity average In to perform an IT statistics step 250 to record the intensity mean In according to the time and generate a statistical value 250A. The control module 160 can perform a statistical value 250A. Step 260 is interpreted to generate a read signal 260A.

Referring to the third and fourth figures, in accordance with a preferred embodiment of the present invention, the image capturing module 120 further includes a light source projection device 110 and an image receiving device 130, wherein the light intensity image detection device The system 100 projects a light source outside the light intensity image detecting system 100 by the light source projection device 110 of the image capturing module 120, such as a light emitting diode element (LED), and the image capturing module 120 is used by the image capturing module 120. The image receiving device 130 receives the light source reflection to generate the first image data 210A.

Referring to the third and fourth figures, in accordance with a preferred embodiment of the present invention, the information processing module 140 further includes a grayscale unit 145, a halo quantification unit 150, and a pixel calculation unit 155. The grayscale unit 145 receives the first image data 210A and performs a grayscale step 220 to convert the color information of the first image data 210A to generate a second image data 220A, thereby greatly reducing the image information. Volume and speed up the next steps. In addition, the halo quantification unit 150 receives the second image data 220A to perform a halo calculation step 230, thereby determining the halo range (ROI) of the image while removing the non-halo area to generate a third image data 230A. . Furthermore, the third image data 230A is received by the pixel calculation unit 155 to perform an intensity calculation step 240. The intensity calculation step 240 can calculate the average of the intensities of all the pixels of the third image data 230A and generate an intensity. Mean In.

Referring to the third and fourth figures, in accordance with a preferred embodiment of the present invention, the control module 160 further includes an IT computing unit 165 and a control unit 170, wherein the IT computing unit 165 records the time according to the time. The information processing module 140 outputs the intensity average In over time and performs an IT statistics step 250 to record the intensity mean In according to time and generate a statistical value 250A, as shown in the fifth figure. In addition, the control unit 170 receives the statistical value 250A to perform the interpretation step 260 and generates a interpretation signal 260A. When the In value of the statistical value becomes stronger with time, the interpretation signal 260A indicates that the image is close to the light intensity type image detection. The measurement system 100, if the In value of the statistical value becomes weaker with time, the interpretation signal 260A indicates that the image is far away from the light intensity type image detection system 100, and if the In value of the statistical value does not continuously increase/decrement with time, then The interpretation signal 260A indicates that there is no object in front of the light intensity image detection system 100.

Referring to the third and fourth figures, in accordance with a preferred embodiment of the present invention, the light intensity image detection system 100 further includes an operation module 180 having a power unit 185 and a transmission. The unit 190 is, for example, a power unit 185 such as a general motor, a servo motor or a stepping motor. The control unit 170 can perform a control step 270 to generate a control signal 270A according to the interpretation signal 260A, and the operation module 180 can receive the control signal 270A to operate and control the power unit 185, thereby driving the operation of the transmission unit 190.

Referring to the third and fourth figures, in accordance with a preferred embodiment of the present invention, the present invention provides a light intensity image detection method, which first provides a light intensity image detection system 100, and then activates an image capture mode. The light source projection device 110 of the group 120 performs a light source projection step 200 to project a light source outside the light intensity image detection system 100, and performs an image capture step 210 by the image receiving device 130 of the image capture module 120 to receive The light source reflects and produces a first image data 210A. Then, the first image data 210A is received by the information processing module 140 and transmitted to the grayscale unit 145 to perform a grayscale step 220, so as to convert the color information of the first image data 210A to generate a second image data 220A. This can greatly reduce the amount of image information and speed up the next steps. Then, the halo calculating unit 150 performs a halation calculation step 230 according to the second image data 220A to determine a halo range (ROI) of the image, while removing the non-halo area and generating a third image data 230A. Then, the intensity calculation step 240 is performed by the pixel calculation unit 155 according to the third image data 230A to calculate the average of the intensities of all the pixels of the third image data 230A and generate an intensity mean In.

Referring to the third and fourth embodiments, in accordance with a preferred embodiment of the present invention, the information processing module 140 outputs the intensity average In to the control module 160 to facilitate the IT computing unit 165 to record the information processing module 140 according to the time. The intensity mean In is continuously generated over time and an IT statistics step 250 is performed thereby generating a statistical value 250A. Thereafter, the statistical value 250A is received by the control unit 170 to perform the interpretation step 260 and generate a interpretation signal 260A. When the In value of the statistical value becomes stronger with time, the interpretation signal 260A indicates that the image is close to the light intensity image. The detection system 100, if the In value of the statistical value becomes weaker with time, the interpretation signal 260A indicates that the image is far away from the light intensity type image detection system 100, and if the In value of the statistical value does not continuously increase/decrement with time, The interpretation signal 260A indicates that there is no object in front of the light intensity image detection system 100. The control module 160 performs the control step 270 according to the interpretation signal 260A by the control unit 170 to generate the control signal 270A. Thereafter, the control module 160 outputs the control signal 270A to the operation module 180 to operate and control the power unit 185, thereby driving the operation of the transmission unit 190.

Obviously, the present invention may have many modifications and differences as described in the above embodiments. It is therefore to be understood that within the scope of the appended claims, the invention may be The above are only the preferred embodiments of the present invention, and are not intended to limit the scope of the present invention; any equivalent changes or modifications made without departing from the spirit of the present invention should be included in the following patent application. Within the scope.

100. . . Light intensity image detection system

110. . . Light source projection device

120. . . Image capture module

130. . . Image receiving device

140. . . Information processing module

145. . . Grayscale unit

150. . . Halo quantification unit

155. . . Pixel calculation unit

160. . . Control module

165. . . I-T calculation unit

170. . . control unit

180. . . Operation module

185. . . Power unit

190. . . Transmission unit

200. . . Light source projection step

210. . . Image capture step

210A. . . First image data

220. . . Grayscale step

220A. . . Second image data

230. . . Halo calculation step

230A. . . Third image data

240. . . Intensity calculation step

In. . . Intensity mean

250. . . I-T statistical steps

250A. . . Statistical value

260. . . Interpretation step

260A. . . Interpretation signal

270. . . Control step

270A. . . Control signal

The first figure is a schematic diagram of the halo principle of the present invention;

The second figure is a schematic diagram illustrating the halation intensity of the present invention;

The third figure is a schematic diagram of the intensity image detection system of the present invention;

The fourth figure is a schematic diagram of the intensity image detection method of the present invention;

The fifth figure shows the I-T statistical diagram of the present invention.

100. . . Light intensity image detection system

110. . . Light source projection device

120. . . Image capture module

130. . . Image receiving device

140. . . Information processing module

145. . . Grayscale unit

150. . . Halo quantification unit

155. . . Pixel calculation unit

160. . . Control module

165. . . I-T calculation unit

170. . . control unit

180. . . Operation module

185. . . Power unit

190. . . Transmission unit

Claims (20)

A light intensity image detection system includes: an image capture module, wherein the image capture module is configured to perform an image capture step to generate a first image data; a processing module, the information processing module receives the first image data to perform an intensity calculation step and generates an intensity mean value; and a control module, the control module further includes an IT calculation unit to receive the intensity mean value and perform An IT statistical step of generating a statistical value, wherein the IT statistical step is that the IT computing unit records the intensity average value output by the information processing module over time according to time; wherein the control module further includes a control unit. The control unit receives the statistical value to perform an interpretation step and generates the interpretation signal, and the control unit performs a control step according to the interpretation signal to generate a control signal; wherein the determining step is determined by the control unit according to the statistical value Whether the image is close to the light intensity type image detecting system; wherein the above-mentioned interpretation step interprets the intensity mean value When the time becomes stronger, the interpretation signal indicates that the image is close to the light intensity image detection system; wherein the interpretation step determines that the intensity average becomes weaker with time, the interpretation signal indicates that the image is away from the light intensity image detection. Measurement system The above-mentioned interpretation step determines that the intensity mean does not continuously increase/decrement with time, and the interpretation signal indicates that the light intensity image detection system has no image before. The light intensity type image detecting system of claim 1, wherein the image capturing module further comprises a light source projection device for projecting a light source outside the light intensity image detecting system. The light intensity type image detecting system of claim 2, wherein the light source projection device further comprises a light emitting diode element. The light intensity type image detecting system of claim 1, wherein the image capturing module further comprises an image receiving device for receiving light source reflection and generating the first image data. The light intensity image detection system of claim 1, wherein the information processing module further comprises: a grayscale unit, the grayscale unit receives the first image data to generate a second Image data; a halo quantifying unit, the halo quantifying unit receives the second image data to generate a third image data; and a pixel computing unit, the pixel calculating unit receives the third image Data to generate this intensity mean. The light intensity image detection system of claim 5, wherein the grayscale unit performs a grayscale step to convert the color information of the first image data into the second image data. The light intensity type image detecting system of claim 5, wherein the halo quantifying unit performs a halo calculating step to determine a halo range of the image while removing the non-halo area to form the third video material. The light intensity type image detecting system of claim 5, wherein the pixel calculation unit performs the intensity calculation step to calculate an average value of the intensities of all pixels of the third image data and generate the intensity mean value. . The light intensity image detection system of claim 1, wherein the light intensity image detection system further comprises an operation module, and the operation module receives the control signal. The light intensity image detection system of claim 1, wherein the operation module further comprises a power unit and a transmission unit, wherein the operation module receives the control signal to operate and control the power The unit, thereby driving the actuation of the transmission unit. The light intensity type image detecting system of claim 1, wherein the power unit further comprises a motor. A light intensity image detecting method, comprising: performing an image capturing step to receive a light source reflection and generating a first image data; performing a grayscale step to convert the first image data The color information further generates a second image data; performing a halation calculation step according to the second image data to determine a halo range of the image, and simultaneously removing the non-halo area and generating a third image data; and according to the third The image data is subjected to an intensity calculation step to calculate a mean value of the intensities of all the pixels of the third image data and generate an intensity mean value, wherein the intensity mean value is continuously generated with time; the intensity mean value outputted with time according to the time record Perform an IT statistical step and generate a statistical value; The statistical value is used to perform a reading process, and a reading signal is generated, wherein the intensity average of the statistical value becomes stronger with time, the reading signal indicates that the image is close, and the intensity average of the statistical value is As the time becomes weaker, the interpretation signal indicates that the image is far away, and the intensity average of the statistical value does not change with time, the interpretation signal indicates no image. The light intensity type image detecting method according to claim 12, wherein the light intensity type image detecting method is performed by the light intensity type image detecting system according to claim 1 of the patent application. The light intensity image detecting method according to claim 12, wherein the light intensity image detecting method further comprises a light source projecting step of projecting a light source by a light source projecting device. The light intensity type image detecting method according to claim 12, wherein the image capturing step is performed by an image receiving device receiving light source reflection and generating the first image data. Light intensity image detection method as described in claim 12 In the above method, the grayscale step is performed by converting a color information of the first image data by a grayscale unit to generate the second image data. The light intensity type image detecting method according to claim 12, wherein the halo calculating step determines the halo range of the image according to the second image data by the halo quantifying unit, and removes the non-light The halo region produces the third image data. The light intensity image detecting method according to claim 12, wherein the intensity calculating step calculates the intensity of all pixels of the third image data according to the third image data by a pixel computing unit. The mean and produce the intensity mean. The light intensity image detection method according to claim 13 , wherein the light intensity image detection method generates a control signal according to the interpretation signal to perform a control step to control the light intensity image detection system. Follow-up action. The light intensity image detecting method according to claim 19, wherein the control signal is used to control a power unit to drive a driving unit.