A working image transmission method for gravure printing machine

Abstract

The present invention relates to the field of image transmission technology, and in particular to a working image transmission method for a gravure printing machine, the method comprising: for any frame of working image in a working video, calculating the consistency of detail features with each preset direction, and then determining the target direction, obtaining a lossy image through reconstruction and upsampling according to a downsampled sub-image of any resolution and a detail sub-image of the target direction, calculating compression distortion according to the area of ​​the overlapped area of ​​the motion area of ​​the working image and the lossy image and the difference in the gradient amplitude and gradient direction of the pixel points in the overlapped area, calculating the preference of any resolution according to the compression distortion and compression rate, compressing the downsampled sub-image of the resolution with the largest preference and the detail sub-image of the target direction as the lossy compression result of the working image, so as to facilitate transmission. The present invention retains the rich time domain information in the working video and the detail information of the motion area.

Description

Technical Field

[0001] The present invention relates to the technical field of image transmission, and more particularly to a working image transmission method for an intaglio printing machine. Background Art

[0002] Gravure printing presses can provide high-quality printing at a high speed. With the improvement of automation, the control and monitoring system of gravure printing presses can automatically record various operating data and work videos. By transmitting the operating data and work videos to the ERP (Enterprise Resource Planning) system, it is convenient for managers to grasp the working status of the printing press in real time, and then carry out production management and implementation of production plans.

[0003] In the related art, for example, a Chinese patent application document with application publication number CN115239825A discloses a solar cell appearance inspection image compression method based on wavelet transform, which includes the following steps: (1) The application uses a haar wavelet function with better compression decomposition to decompose and obtain a low-frequency component A, a horizontal high-frequency component H, a vertical high-frequency component V and a diagonal high-frequency component D; (2) Since the important and useful information after wavelet decomposition is distributed in the low-frequency component, the high-frequency coefficient is threshold quantized each time; (3) After the signal is reduced after denoising, the signal is reduced and the interference of noise in the image is reduced. Then, an inverse wavelet transform is used to greatly reduce the size of the image with basically no distortion; (4) The compressed image is judged for appearance through an image analysis algorithm.

[0004] This application decomposes the image through Haar wavelet transform and performs threshold quantization on the high-frequency coefficients obtained by the decomposition. The compressed image has less loss than the original image and reduces storage space. However, Haar wavelet transform mainly focuses on spatial domain features. When processing dynamic scenes such as the working video of a gravure printing machine that contain rich time domain information, it lacks effective acquisition of motion features, which leads to a significant decrease in the quality of the compressed video. Summary of the invention

[0005] In order to solve the technical problem that when processing the working video of a gravure printing press, which contains rich time domain information, by Haar wavelet transform, there is a lack of effective acquisition of motion features, which leads to a significant decrease in the quality of the compressed video, the present invention provides a working image transmission method for a gravure printing press, comprising: acquiring a working video of the gravure printing press; for any frame of working image in the working video, calculating the consistency of the detail features of the working image with each preset direction, wherein the consistency is negatively correlated with the angle between the gradient direction of the pixel point in the working image and the preset direction, and positively correlated with the gradient amplitude of the pixel point; taking the preset direction with the greatest consistency as the target direction; decomposing the working image by wavelet transform to obtain down-sampled sub-images of different resolutions and detail sub-images of each preset direction. ; For the downsampled sub-image of any resolution and the detail sub-image of the target direction, reconstruct an intermediate image based on the downsampled sub-image and the detail sub-image in the target direction; upsample the intermediate image to obtain a lossy image of the same size as the working image; obtain the motion areas of the working image and the lossy image by the inter-frame difference method; calculate the compression distortion, which is negatively correlated with the area of ​​the overlapping area of ​​the two motion areas, and positively correlated with the difference in gradient amplitude and gradient direction of the pixel points in the overlapping area; calculate the preference of any resolution, which is negatively correlated with the compression distortion and positively correlated with the compression ratio; the compression ratio is equal to the ratio of the resolution to the size of the working image; compress the downsampled sub-image of the resolution with the largest preference and the detail sub-image in the target direction by a lossless compression algorithm as the lossy compression result of the working image.

[0006] The present invention obtains the target direction through the consistency of the detail features of the working image with each preset direction, obtains the lossy image through reconstruction and upsampling according to the down-sampled sub-image and the detail sub-image of the target direction, removes redundant information, improves the efficiency of image compression and transmission, and at the same time enables the detail features in the working image to be best preserved in the lossy image; according to the size of the area of ​​the overlapping area of ​​the motion area of ​​the working image and the lossy image and the difference in the gradient amplitude and gradient direction of the pixel points in the overlapping area, the compression distortion is calculated, the rich time domain information in the working video of the gravure printing machine and the detail information of the motion area are preserved, the quality of the compressed video is guaranteed, and the technicians can better identify and solve printing defects, thereby ensuring the quality of the final product.

[0007] Preferably, the consistency between the detail features of the working image and each preset direction satisfies the expression: ; In the formula, Detailed features and preset directions for working images consistency, For the preset direction, , They are the first The gradient direction and gradient amplitude of each pixel, is the gradient direction With preset direction The angle of is the cosine function, is the number of all pixels in the working image.

[0008] The present invention determines the target direction through the consistency between the detail features of the working image and each preset direction, so that the detail features in the working image can be best preserved and the loss of important information can be reduced.

[0009] Preferably, the preset directions include a horizontal direction, a diagonal direction and a vertical direction.

[0010] Preferably, reconstructing the intermediate image according to the down-sampled sub-image and the detail sub-image of the target direction comprises: the resolution of the reconstructed intermediate image is ,in, is the resolution of the down-sampled sub-image and the detail sub-image in the target direction; and a 2×2 image block in the intermediate image is reconstructed according to the grayscale value of any pixel in the down-sampled sub-image and the detail sub-image in the target direction.

[0011] Preferably, the method for acquiring the image block comprises: in response to the target direction being a horizontal direction, ; ; In response to the target direction being vertical, ; ; In response to the target direction being a diagonal direction, ; ; In the formula, , are the grayscale values ​​of the pixels in the downsampled sub-image and the detail sub-image in the target direction, respectively; , , , They are the grayscale values ​​of the pixel in the upper left corner, the pixel in the upper right corner, the pixel in the lower left corner, and the pixel in the lower right corner of the image block respectively.

[0012] Preferably, the calculating compression distortion comprises: ; In the formula, For compression distortion, is the area of ​​the overlapped area of ​​the two motion areas, and is equal to the number of pixels in the overlapped area of ​​the two motion areas. is the size of the working image and is equal to the number of all pixels in the working image. , They are the first The gradient direction and gradient amplitude of each pixel, , are the first The gradient direction and gradient amplitude of each pixel, The first The gradient direction of the pixel is consistent with the gradient direction of the first pixel in the motion area of ​​the lossy image. The angle between the gradient directions of the pixels, is a sine function, is the absolute value function.

[0013] The present invention can help quantify the degree of retention of motion information during the compression process by calculating the area of ​​the overlapping area between the motion area in the working image and the motion area in the compressed lossy image, and can help quantify the loss of motion information during the compression process by calculating the difference in gradient amplitude and direction of pixel points in the overlapping area.

[0014] Preferably, the preference satisfies the expression: ; In the formula, For the preference, For compression distortion, is the compression ratio.

[0015] The present invention calculates the optimization degree of each resolution through compression distortion and compression rate, and can ensure that the lossy compression result of the working image finally obtained has high compression rate and low compression distortion.

[0016] Preferably, the preference satisfies the expression: ; In the formula, For the preference, For compression distortion, is the compression ratio, , are respectively the preset first weight and the second weight.

[0017] When calculating the priority, the present invention sets different weights for the compression rate and the compression distortion, and can adjust the compression strategy according to the specific needs of the administrator, so as to better adapt to different usage scenarios.

[0018] Preferably, the transmission includes: combining the lossy compression results of all working images in the working video to form a lossy compression result of the working video; and transmitting the lossy compression result of the working video to the ERP system through the communication module.

[0019] The beneficial effects of the present invention are:

[0020] The present invention obtains the target direction through the consistency of the detail features of the working image with each preset direction, obtains the lossy image through reconstruction and upsampling according to the down-sampled sub-image and the detail sub-image of the target direction, removes redundant information, improves the efficiency of image compression and transmission, and at the same time enables the detail features in the working image to be best preserved in the lossy image; according to the size of the area of ​​the overlapping area of ​​the motion area of ​​the working image and the lossy image and the difference in the gradient amplitude and gradient direction of the pixel points in the overlapping area, the compression distortion is calculated, the rich time domain information in the working video of the gravure printing machine and the detail information of the motion area are preserved, the quality of the compressed video is guaranteed, and the technicians can better identify and solve printing defects, thereby ensuring the quality of the final product. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below through the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0022] Figure 1 is a flow chart schematically illustrating a working image transmission method for a gravure printing press in the present invention;

[0023] Figure 2 FIG. 1 schematically illustrates the process of decomposing a working image by wavelet transform. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] The embodiment of the present invention discloses a working image transmission method for a gravure printing machine, referring to Figure 1 , including steps S1 to S6:

[0027] S1. Collect working video of gravure printing machine.

[0028] It should be noted that the gravure printing machine can provide high-quality printing at a relatively high speed. With the improvement of the degree of automation, the control and monitoring system of the gravure printing machine can automatically record various operating data and working videos. By transmitting the operating data and working videos to the ERP (Enterprise Resource Planning) system, it is convenient for managers to grasp the working status of the printing machine in real time, and then carry out production management and implementation of production plans. The ERP system is an integrated management system that realizes information sharing and resource optimization by integrating the company's financial, human resources, production, supply chain and other business processes.

[0029] Specifically, a plurality of cameras are arranged on the body of the gravure printing machine, and a communication module is added to the control system of the gravure printing machine; the cameras are used to collect working videos of the gravure printing machine, and the communication module is used to lossily compress the collected working videos, and transmit the lossy compression results of the working videos to the ERP system, so that the management personnel can grasp the working status of the printing machine in real time, and then carry out production management and implementation of production plans.

[0030] Among them, the resolutions of common cameras include 1920px×960px (i.e. 960P), 1920px×1080px (i.e. 1080P), 3840px×2160px (i.e. 4K ultra-high definition), etc. px (pixel) is the smallest image unit; in the two sets of numbers in imaging, the former represents the length of the image in the captured video, and the latter represents the width of the image in the captured video. The product of the two is the resolution of the image in the captured video. For example, in 1920px×960px, 1920px represents the length of the image in the captured video, and 960px represents the width of the image in the captured video.

[0031] S2. Decompose the working image by wavelet transform to obtain down-sampled sub-images of different resolutions and detail sub-images of each preset direction.

[0032] Specifically, the working image is decomposed into multiple levels through Haar wavelet transform, and different levels of decomposition obtain down-sampled sub-images of different resolutions and detail sub-images in each preset direction, wherein the preset directions include horizontal, diagonal and vertical directions; wherein each level of decomposition decomposes the down-sampled sub-image obtained by the previous level of decomposition, mainly by low-pass filtering and high-pass filtering in the horizontal and vertical directions, and as the decomposition level increases, the resolution of the obtained down-sampled sub-image and the detail sub-image in each preset direction becomes lower; Haar wavelet transform is a well-known technology and will not be described in detail here.

[0033] For example, the process of decomposing the working image by wavelet transform is as follows: Figure 2As shown, sub-image LL1 is a down-sampled sub-image obtained under the first-level decomposition. The down-sampled sub-image is the result of downsampling the working image, and contains the low-frequency components of the working image in the horizontal and vertical directions; sub-image LH1 is a detail sub-image in the vertical direction obtained under the first-level decomposition. The detail sub-image in the vertical direction contains the low-frequency components of the working image in the horizontal direction and the high-frequency components in the vertical direction. Therefore, the detail sub-image in the vertical direction contains the detail features of the working image in the vertical direction; sub-image HL1 is a detail sub-image in the horizontal direction obtained under the first-level decomposition. The detail sub-image in the horizontal direction contains the high-frequency components of the working image in the horizontal direction and the low-frequency components in the vertical direction. Therefore, the detail sub-image in the horizontal direction contains the detail features of the working image in the horizontal direction; sub-image HH1 is a detail sub-image in the diagonal direction obtained under the first-level decomposition. The detail sub-image in the diagonal direction contains the high-frequency components of the working image in the horizontal direction and the high-frequency components in the vertical direction. Therefore, the detail sub-image in the diagonal direction contains the detail features of the working image in the diagonal direction.

[0034] Since the present invention finally selects a down-sampled sub-image of a certain resolution and a detail sub-image of the target direction used to obtain the lossy compression result of the working image according to the compression distortion and compression rate, and the resolution of the obtained sub-image becomes lower as the decomposition level increases, therefore, in order to ensure the image quality of the lossy compression result, it is necessary to limit the decomposition level of the wavelet transform to not be too large; in this embodiment, the decomposition level is 4, and in other embodiments, the decomposition level can be set according to the actual application scenario and requirements.

[0035] Among them, when the working image is multi-level decomposed by Haar wavelet transform, the grayscale value of each pixel in the down-sampled sub-image, the detail sub-image in the vertical direction, the detail sub-image in the horizontal direction, and the detail sub-image in the diagonal direction is calculated according to the grayscale value of all pixels in each image block of size 2×2 of the working image, specifically:

[0036] ;

[0037] ;

[0038] ;

[0039] ;

[0040] In the formula, , , , They are the grayscale values ​​of the pixels in the downsampling sub-image, the vertical detail sub-image, the horizontal detail sub-image, and the diagonal detail sub-image, respectively. , , , They are respectively the grayscale values ​​of the pixel in the upper left corner, the pixel in the upper right corner, the pixel in the lower left corner, and the pixel in the lower right corner of the image block of size 2×2 in the working image.

[0041] S3. Determine the target direction according to the consistency between the detailed features of the working image and each preset direction.

[0042] It should be noted that the detailed features in the image (such as texture, color, shape, edge, etc.) are the basis for analyzing and identifying the image content. These features can reflect the local or overall structure of the image. Therefore, it is expected that the detailed features will be retained in the lossy compression result of the final working image.

[0043] Specifically, the detail features in the working image are judged by the gradient of the pixel points. Therefore, for any frame of the working image in the working video, edge detection is performed on the working image to obtain the gradient amplitude and gradient direction of each pixel point in the working image; the algorithms used for edge detection include but are not limited to the Canny algorithm, the Sobel algorithm and the Laplace algorithm. The Canny algorithm, the Sobel algorithm and the Laplace algorithm are all well-known technologies and will not be described in detail here.

[0044] Further, the consistency between the detail features of the working image and each preset direction is calculated, and the preset directions include a horizontal direction, a diagonal direction, and a vertical direction; the consistency between the detail features of the working image and each preset direction satisfies the expression:

[0045] ;

[0046] In the formula, Detailed features and preset directions for working images consistency, For the preset direction, , They are the first The gradient direction and gradient magnitude of each pixel, is the gradient direction With preset direction The angle of is the cosine function, is the number of all pixels in the working image.

[0047] The larger the gradient amplitude of the pixel in the working image, the more the pixel belongs to the detail feature in the working image; the smaller the angle between the gradient direction of the pixel and the preset direction, the more the detail feature in the working image conforms to the preset direction. Therefore, the consistency is negatively correlated with the angle between the gradient direction of the pixel point in the working image and the preset direction, and positively correlated with the gradient amplitude of the pixel point.

[0048] Furthermore, the preset direction with the greatest consistency is taken as the target direction.

[0049] It should be noted that by determining the target direction, the detailed features in the working image can be best preserved and the loss of important information can be reduced, which is very important for image compression and transmission.

[0050] S4. According to the downsampled sub-image of any resolution and the detail sub-image of the target direction, a lossy image is obtained by reconstruction and upsampling, and the compression distortion is calculated according to the area of ​​the overlapping area of ​​the motion area of ​​the working image and the lossy image and the difference in the gradient amplitude and gradient direction of the pixel points in the overlapping area.

[0051] It should be noted that the Haar wavelet transform mainly processes images in the spatial domain, extracting low-frequency and high-frequency components by decomposing the image. In dynamic videos, the changes in the moving areas are frequent and complex, resulting in limited effectiveness of the static Haar wavelet transform in processing motion information. Motion information is key in dynamic videos and can help identify changes between frames in the video. If the compression result lacks effective motion information, the details of the moving areas will be lost when the video is compressed, resulting in a significant decrease in the quality of the compressed video.

[0052] It should be further explained that the present invention can help quantify the degree of retention of motion information during the compression process by calculating the area of ​​the overlapping area of ​​the motion area in the working image and the motion area in the compressed lossy image, and the larger the area of ​​the overlapping area, the better the performance of the compressed lossy image in retaining motion information; the change in gradient amplitude can indicate the degree of loss of image details, and the difference in gradient direction can reflect the change in motion direction and speed. Calculating the difference in gradient amplitude and direction of pixel points in the overlapping area can help quantify the loss of motion information during the compression process, and the gradient amplitude and direction of the overlapping area remain similar, indicating that the compression process retains the motion area well, otherwise, it indicates that the distortion is serious after compression.

[0053] Specifically, when the intermediate image is reconstructed based on the down-sampled sub-image and the detail sub-image of the target direction, the resolution of the reconstructed intermediate image is ,in, is the resolution of the downsampled sub-image and the detail sub-image in the target direction, specifically: according to the grayscale value of any pixel in the downsampled sub-image and the detail sub-image in the target direction, a 2×2 image block in the intermediate image is reconstructed, and the image block includes a pixel in the upper left corner, a pixel in the upper right corner, a pixel in the lower left corner, and a pixel in the lower right corner.

[0054] The grayscale values ​​of the four pixels in the image block are obtained as follows:

[0055] In response to the target direction being horizontal, ; ;

[0056] In response to the target direction being the vertical direction, ; ;

[0057] In response to the target direction being a diagonal direction, ; ;

[0058] In the formula, , are the grayscale values ​​of the pixels in the downsampled sub-image and the detail sub-image in the target direction, respectively; , , , They are the grayscale values ​​of the pixel in the upper left corner, the pixel in the upper right corner, the pixel in the lower left corner, and the pixel in the lower right corner of the image block respectively.

[0059] Furthermore, the intermediate image is upsampled to obtain a lossy image of the same size as the working image; the algorithms used for the upsampling include but are not limited to the nearest neighbor interpolation algorithm, the bilinear interpolation algorithm, and the bicubic interpolation algorithm. The nearest neighbor interpolation algorithm, the bilinear interpolation algorithm, and the bicubic interpolation algorithm are all well-known technologies and will not be described in detail here.

[0060] Furthermore, the motion regions of the working image and the lossy image are obtained by using an inter-frame difference method. The inter-frame difference method is a well-known technology and will not be described in detail here.

[0061] It should be noted that the working video captured by the camera is continuous. If there is a moving target in the working scene, there will be obvious changes between consecutive frames. The temporal difference method draws on this idea. Since the moving target in the working scene is moving, the position of the moving target in different frames of working images is different. The temporal difference method performs differential operations on multiple frames of working images that are continuous in time, and subtracts the corresponding pixels of different frames of working images to determine the moving area, thereby realizing the detection of the moving area.

[0062] Further, the compression distortion is calculated according to the area of ​​the overlapped area of ​​the motion area of ​​the working image and the lossy image and the difference in the gradient amplitude and gradient direction of the pixels in the overlapped area, wherein the area of ​​the overlapped area is equal to the number of pixels in the overlapped area; and the compression distortion satisfies the expression:

[0063] ;

[0064] In the formula, For compression distortion, is the area of ​​the overlapped area of ​​the two motion areas, and is equal to the number of pixels in the overlapped area of ​​the two motion areas. is the size of the working image and is equal to the number of all pixels in the working image. , They are the first The gradient direction and gradient amplitude of each pixel, , are the first The gradient direction and gradient amplitude of each pixel, The first The gradient direction of the pixel is consistent with the gradient direction of the first pixel in the motion area of ​​the lossy image. The angle between the gradient directions of the pixels, is a sine function, is the absolute value function.

[0065] The larger the overlap area of ​​the motion area of ​​the working image and the lossy image, the more complete the motion area of ​​the working image is retained by the obtained lossy image, and the smaller the compression distortion of the lossy image is compared with the working image; The gradient direction of the pixel is consistent with the gradient direction of the first pixel in the motion area of ​​the lossy image. The smaller the angle of the gradient direction of each pixel, the smaller the angle of the gradient direction of each pixel, and the ... The gradient amplitude of the pixel is the same as the gradient amplitude of the first pixel in the motion area of ​​the lossy image. The difference in gradient amplitude of pixels The smaller it is, the more complete the motion area of ​​the working image is retained by the obtained lossy image, and the smaller the compression distortion of the lossy image compared to the working image; the compression distortion is negatively correlated with the area of ​​the overlapping area of ​​the two motion areas, and positively correlated with the difference in gradient amplitude and gradient direction of the pixel points in the overlapping area.

[0066] S5. Calculate the preference of any resolution based on the compression distortion and compression ratio.

[0067] It should be noted that in order to ensure that the lossy compression result of the working image finally obtained has a high compression ratio and low compression distortion, the present invention calculates the preference of any resolution according to the compression distortion and the compression ratio.

[0068] Specifically, the preference of any resolution is calculated, and the preference is negatively correlated with the compression distortion and positively correlated with the compression ratio; wherein the compression ratio is equal to the ratio of the size of the working image to the resolution.

[0069] In one embodiment, the preference satisfies the expression:

[0070] ;

[0071] In the formula, For the preference, For compression distortion, is the compression ratio.

[0072] In another embodiment, the preference satisfies the expression:

[0073] ;

[0074] In the formula, For the preference, For compression distortion, is the compression ratio, is the preset weight.

[0075] It should be noted that in order to ensure that the lossy compression result of the final working image has a high compression rate and low compression distortion at the same time, it is required to calculate the preference of any resolution based on the compression distortion and compression rate. The greater the compression rate and the smaller the compression distortion, the greater the preference. That is, the preference is negatively correlated with the compression distortion and positively correlated with the compression rate.

[0076] Among them, for managers, when reducing the size of the compressed result (i.e., high compression ratio) is more important than maintaining high-quality video (i.e., low compression artifacts), the preset weights are The value range is (0.5,1), for example, the preset weight ,at this time ; When maintaining high video quality (i.e. low compression artifacts) is more important than reducing the size of the compressed result (i.e. high compression ratio), the preset weights The value range is (0,0.5), for example, the preset weight ,at this time .

[0077] By setting different weights, you can adjust the compression strategy according to the specific needs of administrators; when a high compression rate is required in certain scenarios (such as network transmission or limited storage space), you can set the compression rate weight. Greater than the weight of compression distortion In other cases, image quality may be more important, so you can set the weight of the compression rate. Less than the weight of compression distortion ; Can better adapt to different usage scenarios.

[0078] S6. Compress the downsampled sub-image with the highest resolution and the detail sub-image in the target direction by using a lossless compression algorithm as a lossy compression result of the working image for easy transmission.

[0079] Specifically, the downsampled sub-image with the highest preferred resolution and the detail sub-image in the target direction are compressed by a lossless compression algorithm as a lossy compression result of the working image; the lossless compression algorithm includes but is not limited to Huffman coding, arithmetic coding, etc.

[0080] Furthermore, the lossy compression results of all working images in the working video are combined to form a lossy compression result of the working video; and the lossy compression result of the working video is transmitted to the ERP system through the communication module.

[0081] The present invention obtains the target direction through the consistency of the detail features of the working image with each preset direction, obtains the lossy image through reconstruction and upsampling according to the down-sampled sub-image and the detail sub-image of the target direction, removes redundant information, improves the efficiency of image compression and transmission, and at the same time enables the detail features in the working image to be best preserved in the lossy image; according to the size of the area of ​​the overlapping area of ​​the motion area of ​​the working image and the lossy image and the difference in the gradient amplitude and gradient direction of the pixel points in the overlapping area, the compression distortion is calculated, the rich time domain information in the working video of the gravure printing machine and the detail information of the motion area are preserved, the quality of the compressed video is guaranteed, and the technicians can better identify and solve printing defects, thereby ensuring the quality of the final product.

[0082] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0083] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A working image transmission method for a gravure printing press, characterized in that: include: Collect working videos of gravure printing machines; For any frame of working image in the working video, the consistency of the detail features of the working image and each preset direction is calculated. The consistency is negatively correlated with the angle between the gradient direction of the pixel point in the working image and the preset direction, and positively correlated with the gradient amplitude of the pixel point; the preset direction with the greatest consistency is taken as the target direction; the working image is decomposed by wavelet transform to obtain downsampled sub-images of different resolutions and detail sub-images of each preset direction; For the downsampled sub-image of any resolution and the detail sub-image of the target direction, an intermediate image is reconstructed according to the downsampled sub-image and the detail sub-image of the target direction. The resolution of the intermediate image is ,in, is the resolution of the down-sampled sub-image and the detail sub-image in the target direction; reconstructing a 2×2 image block in the intermediate image according to the grayscale value of any pixel in the down-sampled sub-image and the detail sub-image in the target direction, including: in response to the target direction being the horizontal direction, ; ; In response to the target direction being vertical, ; ; In response to the target direction being a diagonal direction, ; ; In the formula, , are the grayscale values ​​of the pixels in the downsampled sub-image and the detail sub-image in the target direction, respectively. , , , The gray values ​​of the upper left pixel, upper right pixel, lower left pixel and lower right pixel in the image block are respectively; the intermediate image is upsampled to obtain a lossy image of the same size as the working image; the motion area of ​​the working image and the lossy image is obtained by the inter-frame difference method; the compression distortion is calculated , ; is the area of ​​the overlapped area of ​​the two motion areas, and is equal to the number of pixels in the overlapped area of ​​the two motion areas. is the size of the working image and is equal to the number of all pixels in the working image. , They are the first The gradient direction and gradient magnitude of each pixel, , are the first The gradient direction and gradient magnitude of each pixel, The first The gradient direction of the pixel is consistent with the gradient direction of the first pixel in the motion area of ​​the lossy image. The angle between the gradient directions of the pixels, is a sine function, is the absolute value function; Calculating the preference of any resolution, wherein the preference is negatively correlated with compression distortion and positively correlated with compression ratio; the compression ratio is equal to the ratio of the size of the working image to the resolution; The down-sampled sub-image with the highest preferred resolution and the detail sub-image in the target direction are compressed by a lossless compression algorithm as the lossy compression result of the working image for transmission.

2. A working image transmission method for a gravure printing press according to claim 1, characterized in that: The consistency between the detail features of the working image and each preset direction satisfies the expression: ； In the formula, Detailed features and preset directions for working images consistency, For the preset direction, , They are the first The gradient direction and gradient magnitude of each pixel, is the gradient direction With preset direction The angle of is the cosine function, is the number of all pixels in the working image.

3. A working image transmission method for a gravure printing press according to claim 1, characterized in that: The preset directions include a horizontal direction, a diagonal direction and a vertical direction.

4. A working image transmission method for a gravure printing press according to claim 1, characterized in that: The preference satisfies the expression: ； In the formula, For the preference, For compression distortion, is the compression ratio.

5. A working image transmission method for a gravure printing press according to claim 1, characterized in that: The preference satisfies the expression: ； In the formula, For the preference, For compression distortion, is the compression ratio, is the preset weight.

6. A working image transmission method for a gravure printing press according to claim 1, characterized in that: The transmission includes: The lossy compression results of all working images in the working video are combined to form a lossy compression result of the working video; and the lossy compression result of the working video is transmitted to the ERP system through the communication module.

Images