CN115564893A - An Image Codec Method Based on Coded Structured Light - Google Patents

Abstract

The invention discloses an image coding and decoding method based on coded structured light. The method of the present invention includes the steps of adopting the method of stripe coding, projecting the pattern in the structured light image into multiple scene images including positive and negative coding patterns; the optimization method based on global illumination reduces the influence of global illumination, and uses Gray code coding Carry out three-dimensional reconstruction of the scene image, eliminate the invalid reconstruction area to obtain the undecoded captured image; perform binarization operation on the captured image of the eliminated invalid reconstruction area to obtain the binary image; superimpose the binary image to construct the decoded image, Then filter and denoise the decoded image to remove noise points in the decoded image. Compared with the prior art, the present invention improves the stability of binarization, removes invalid reconstruction regions, and improves reconstruction precision and encoding and decoding efficiency.

Description

An Image Codec Method Based on Coded Structured Light

technical field

The invention belongs to the technical field of image processing, and in particular relates to an image encoding and decoding method based on encoded structured light.

Background technique

At present, with the development of intelligent manufacturing, non-contact structured light vision sensors are more and more widely used in industrial applications. In the fields of reverse engineering, workpiece quality inspection, and workpiece size measurement, structured light vision sensors have been widely used. . The point cloud reconstruction using the visual sensor of encoded structured light needs to meet the triangulation measurement model. The triangulation measurement model is a non-contact, fast measurement and high precision measurement method. The coded structured light uses a projector to project a special coded pattern onto the object to be measured, and performs three-dimensional reconstruction by collecting the deformed coded pattern, and the changes include the depth information of the surface of the measured object. By decoding and analyzing the collected scene coded image, the decoding value of each pixel is obtained, and the spatial position of the pixel in the image is calculated according to the triangular geometric model formed by the camera and projector, so as to obtain the three-dimensional information of the surface of the measured object. Conventional prior art thinking.

When encoding the structured light system for 3D reconstruction, the encoding and decoding strategy of the structured light projection pattern is an important part. The role of image encoding and decoding is to determine the coordinates of the projection point projected onto the object in the projector image coordinate system and reduce decoding errors. In the prior art, in the reconstruction process of coded structured light, there are global illumination phenomena such as mutual reflection of scene light and subsurface scattering, global illumination will lead to incorrect binary decoding results, and the binarization stability of codec results is poor. The reconstruction of the invalid area affects the encoding and decoding efficiency, and greatly interferes with the reconstruction accuracy.

Contents of the invention

In order to overcome one or more defects and deficiencies in the prior art, the present invention provides an image encoding and decoding method based on encoded structured light, which is used to reduce errors in image decoding and reconstruction.

In order to achieve the above object, the present invention adopts the following technical solutions.

An image encoding and decoding method based on encoded structured light, comprising the following steps:

Using stripe coding, the pattern in the structured light image is projected into multiple scene images including positive coding patterns and reverse coding patterns;

The optimization method based on global illumination reduces the influence of global illumination, uses Gray code encoding to perform three-dimensional reconstruction of scene images, and eliminates invalid reconstruction areas to obtain undecoded captured images;

Performing a binarization operation on the photographed image that eliminates the invalid reconstruction area to obtain a binarized image;

The binary image is superimposed to construct the decoded image, and then the decoded image is filtered and denoised to remove the noise points in the decoded image.

Preferably, when performing stripe coding, vertical stripe coding is used.

Further, when performing stripe coding, 11-bit vertical stripe coding is specifically adopted;

Corresponding to the 11-bit longitudinal stripe code, when projecting the positive coding pattern and the reverse coding pattern, the positive coding pattern projects a total of 11 scene images, and the reverse coding pattern projects a total of 11 scene images, and a total of 22 scenes are projected image.

Further, when performing three-dimensional reconstruction on the scene image, the Gray code code used is a long-bit-width Gray code code.

Further, the process of eliminating invalid reconstruction areas includes:

Obtain the maximum gray value I _max and the minimum gray value I _min of each pixel in all scene images, and then calculate the difference I _dis between the maximum gray value I _max and the minimum gray value I _min , the calculation formula is as follows:

I _dis =I _max -I _min

If I _dis is less than the set threshold, the pixel is located in the invalid reconstruction area, and then the pixel gray value is set to zero, if I _dis is greater than the set threshold, then the pixel is a visible point;

After setting the gray value of all pixels in the scene image whose I _dis is less than the set threshold to zero, the captured image after eliminating the invalid reconstruction area is obtained.

Further, the process of taking an image and performing a binarization operation includes:

Overlap a group of photographed images containing the corresponding positive coding pattern and reverse coding pattern, and then perform binarization judgment, and replace the gray value of the pixel with the judged binarization result to obtain a binarized image.

Further, the process of binarization judgment includes:

为像素点在三维重建后的拍摄图像中反编码图案的灰度值、m表示像素点二值化判断后的结果，则正编码图案、反编码图案进行重合时，像素点的二值化结果的判断公式如下：Let I be the gray value of the positive coding pattern of the pixel in the three-dimensionally reconstructed captured image,

is the gray value of the reverse coding pattern of the pixel in the three-dimensionally reconstructed captured image, and m represents the result of the binarization judgment of the pixel. The judgment formula is as follows:

Wherein, 1 indicates that the pixel point binarization result corresponds to the bright stripes in the longitudinal stripe coding, and 0 indicates that the pixel point binarization result corresponds to the dark stripes in the vertical stripe coding.

Further, the process of superimposing binarized images to construct a decoded image includes:

Superimpose all the binarized images, accumulate the binarized results at the same pixel point on each image to obtain an accumulated value, and then set the accumulated value as a gray value to obtain a decoded image.

Further, the way of filtering and denoising the decoded image is specifically median filtering.

Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

Through the optimization of positive coding patterns, reverse coding patterns, and coding stripes, the stability of binarization is improved, and invalid reconstruction areas are removed; noise is removed from decoded images by median filtering, which reduces errors in decoding and reconstruction. The accuracy of reconstruction and the efficiency of encoding and decoding are improved, and the scope of application of structured light sensors and point cloud reconstruction is expanded.

Description of drawings

FIG. 1 is a general flow chart of one of the image encoding and decoding methods based on encoded structured light in the present invention;

Fig. 2 is the effect diagram when projecting positive coding pattern;

Fig. 3 is an effect diagram when projecting an inverse encoding pattern;

Fig. 4 is the effect figure of binarized image;

FIG. 5 is an effect diagram before filtering and denoising the decoded image;

FIG. 6 is an effect diagram after filtering and denoising the decoded image.

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and the embodiments thereof. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

Example

As shown in Figure 1, an image encoding and decoding method based on encoded structured light in this embodiment, the specific steps are as follows:

S1. Using the method of longitudinal stripe coding, the pattern in the structured light image is projected into multiple scene images including positive and negative coding patterns;

In this embodiment, it is preferable to use 11-bit longitudinal stripe coding during projection, so the positive coding pattern needs to project a total of 11 scene images, and the reverse coding pattern also needs to project a total of 11 scene images, and a total of 22 scene images are obtained by projection;

S2. The optimization method based on global illumination reduces the influence of global illumination, uses Gray code encoding with a longer bit width to perform three-dimensional reconstruction on the scene image in step S1, increases the minimum width of the encoding stripes, and then eliminates invalid reconstruction areas to obtain undecoded capture images;

In the process of 3D reconstruction, since the illumination of the projector has almost no effect on the gray value of the invalid area, the invalid reconstruction area can be eliminated directly according to the change of the gray value of the captured image;

The process of eliminating invalid reconstruction areas includes:

Obtain the maximum gray value I _max and the minimum gray value I _min of each pixel in all scene images, and then calculate the difference I _dis between the maximum gray value I _max and the minimum gray value I _min , the calculation formula is as follows:

I _di s = I _max -I _min

If I _dis is less than the set threshold, the pixel is located in the invalid reconstruction area, and then the gray value of the pixel is set to zero; if I _dis is greater than the set threshold, the pixel is a visible point;

After setting the gray value of all pixels whose I _dis is less than the set threshold in the scene image to zero, obtain the captured image after eliminating the invalid reconstruction area;

In this embodiment, it is preferable to perform three-dimensional reconstruction on all the 22 scene images generated in step S1 and eliminate the invalid reconstruction area, so as to remove the field of view and the area blocked by the object in the captured image;

S3. Perform a binarization operation on the photographed image obtained in step 2 to eliminate the invalid reconstruction area to obtain a binarized image; the process of obtaining the binarized image includes:

为像素点在三维重建后的拍摄图像中反编码图案的灰度值、m表示像素点二值化判断后结果，则正、反两个编码图案进行重合时，像素点的二值化结果的判断公式如下：Superimpose a group of photographed images containing the corresponding positive and negative coding patterns to eliminate the invalid reconstruction area and perform binarization judgment, and replace the gray value of the pixel point with the judged binarization result to obtain a binarized image; positive, negative The two anti-coding patterns themselves are a group of binary coding patterns arranged in chronological order, and there are only two kinds of coding patterns, bright coding stripes and dark coding stripes; let I be the gray color of positive coding patterns in the captured image after three-dimensional reconstruction of pixels. degree value,

is the gray value of the reverse coding pattern of the pixel in the three-dimensionally reconstructed captured image, and m represents the result of the binarization judgment of the pixel point. The judgment formula is as follows:

Wherein, 1 indicates that the pixel point binarization result corresponds to the bright stripe in the longitudinal stripe coding, and 0 represents that the pixel point binarization result corresponds to the dark stripe in the vertical stripe coding; this embodiment is preferably after the scene image in the binarization decoding step S2 , to obtain 11 binarized images; as shown in Figure 2 to Figure 4, the projection positive encoding pattern of Figure 2 and the reverse encoding pattern of Figure 3 obtain the binarized image of Figure 4 after the binarization operation;

S4, superimposing the binarized image in step S3 to construct a decoded image, then filtering and denoising the decoded image, removing noise points in the decoded image, and improving the stability of structured light image decoding;

The process of superimposing the binarized images to construct the decoded image is as follows: superimpose all the binarized images, accumulate the binarization results on the pixels at the same position on each image to obtain an accumulated value, and then set the accumulated value to gray value, thereby obtaining a decoded image; the present embodiment preferably constructs a decoded image from the 11 binarized images obtained in step S3;

Filter and denoise the decoded image, specifically using the median filter method to remove the high-frequency salt and pepper noise in the decoded image, thereby retaining the complete pattern edge; as shown in the comparison between Figure 5 and Figure 6, there are There is an obvious difference in salt and pepper noise. Before filtering and denoising, there is high-frequency salt and pepper noise, and the salt and pepper noise after filtering and denoising is significantly lower than that before filtering and denoising.

Compared with the prior art, the image coding and decoding method based on coded structured light in this embodiment has the following beneficial effects:

This embodiment improves the stability of binarization by optimizing the positive coding pattern, reverse coding pattern, and coding stripes, and removes invalid reconstruction areas; the decoded image is denoised by median filtering, which reduces the time required for decoding and reconstruction. The error improves the accuracy of reconstruction and the efficiency of encoding and decoding, and expands the scope of application of structured light sensors and point cloud reconstruction.

The above-mentioned embodiment is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited by the above-mentioned embodiment, and any other changes, modifications, substitutions, combinations, Simplifications should be equivalent replacement methods, and all are included in the protection scope of the present invention.

Claims (9)

1. An image encoding and decoding method based on encoded structured light, characterized in that, comprising the following steps: Using stripe coding, the pattern in the structured light image is projected into multiple scene images including positive coding patterns and reverse coding patterns; The optimization method based on global illumination reduces the influence of global illumination, uses Gray code encoding to perform three-dimensional reconstruction of scene images, and eliminates invalid reconstruction areas to obtain undecoded captured images; Performing a binarization operation on the photographed image that eliminates the invalid reconstruction area to obtain a binarized image; The binarized image is superimposed to construct the decoded image, and then the decoded image is filtered and denoised to remove the noise points in the decoded image. 2. The image encoding and decoding method based on coded structured light according to claim 1, characterized in that, when performing stripe coding, vertical stripe coding is used. 3. The image encoding and decoding method based on encoded structured light according to claim 2, characterized in that, when performing stripe encoding, specifically adopt 11-bit longitudinal stripe encoding; Corresponding to the 11-bit longitudinal stripe code, when projecting the positive coding pattern and the reverse coding pattern, the positive coding pattern projects a total of 11 scene images, and the reverse coding pattern projects a total of 11 scene images, and a total of 22 scenes are projected image. 4 . The image encoding and decoding method based on coded structured light according to claim 3 , wherein, when performing three-dimensional reconstruction on a scene image, the Gray code used is a long-bit-width Gray code. 5. The image encoding and decoding method based on encoded structured light according to claim 4, wherein the process of eliminating invalid reconstruction areas comprises: Obtain the maximum gray value I _max and the minimum gray value I _min of each pixel in all scene images, and then calculate the difference I _dis between the maximum gray value I _max and the minimum gray value I _min , the calculation formula is as follows: Idis=Imax-Imin If I _dis is less than the set threshold, the pixel is located in the invalid reconstruction area, and then the pixel gray value is set to zero, if I _dis is greater than the set threshold, then the pixel is a visible point; After setting the gray values of all pixels in the scene image whose I _dis is smaller than the set threshold to zero, a captured image after eliminating the invalid reconstruction area is obtained. 6. The image encoding and decoding method based on encoded structured light according to claim 5, wherein the process of taking an image and performing a binarization operation comprises: Overlap a group of photographed images containing the corresponding positive coding pattern and reverse coding pattern, and then perform binarization judgment, and replace the gray value of the pixel with the judged binarization result to obtain a binarized image. 7. The image encoding and decoding method based on encoded structured light according to claim 6, wherein the process of binarization judgment comprises: Let I be the gray value of the positive coding pattern of the pixel in the three-dimensionally reconstructed captured image,

is the gray value of the reverse coding pattern of the pixel in the three-dimensionally reconstructed captured image, and m represents the result of the binarization judgment of the pixel. The judgment formula is as follows:

Wherein, 1 indicates that the pixel point binarization result corresponds to the bright stripes in the longitudinal stripe coding, and 0 indicates that the pixel point binarization result corresponds to the dark stripes in the vertical stripe coding. 8. The image encoding and decoding method based on encoded structured light according to claim 7, wherein the process of superimposing binarized images to construct a decoded image comprises: Superimpose all the binarized images, accumulate the binarized results at the same pixel point on each image to obtain an accumulated value, and then set the accumulated value as a gray value to obtain a decoded image. 9 . The image encoding and decoding method based on encoded structured light according to claim 8 , wherein the way of filtering and denoising the decoded image is specifically median filtering.

Images