An infrared image data compression method for infrared imaging simulation system

Abstract

The invention relates to the field of infrared image data compression technology, and proposes an infrared image data compression method for an infrared imaging simulation system, including: using an infrared imaging simulation system to obtain an infrared image and divide the image block, determining the high-frequency information significance and high-frequency area of ​​the image block; determining the coefficient retention number and retention coefficient of the DCT coefficient matrix in the high-frequency area; compressing and decompressing the infrared image, obtaining a reconstructed infrared image and marking high-frequency information pixels, and determining the track retention degree of the track area; iterating the coefficient retention number with a preset step size, obtaining the updated track retention degree corresponding to each iterative value, determining the retention number suitability of the coefficient retention number, and determining the data compression result of the infrared image according to all the retention number suitabilities. The invention aims to improve the accuracy of information in compressed infrared image data.

Description

Technical Field

[0001] The invention relates to the technical field of infrared image data compression, and in particular to an infrared image data compression method for an infrared imaging simulation system. Background Art

[0002] The infrared imaging simulation system can realize infrared imaging, and has the advantages of long range, good anti-interference, strong penetration ability and can work all day. In order to save the infrared images obtained by the infrared imaging simulation system, it is necessary to compress the infrared images. Generally, DCT discrete cosine transform is used to compress infrared images, transform information from time domain to frequency domain, and concentrate the energy of the signal corresponding to the infrared image data in several frequency components.

[0003] However, in the process of compressing infrared images, different selected frequencies have a great influence on the effect of target trajectory extraction in infrared images. Traditional frequency selection methods often make the target trajectory information contained in the compressed infrared image data unclear, affecting the subsequent extraction and use of the compressed infrared image. Summary of the invention

[0004] The present invention provides an infrared image data compression method for an infrared imaging simulation system to solve the problem that the target trajectory information contained in the compressed infrared image data is unclear due to inappropriate frequency selection. The technical solution adopted is as follows:

[0005] An embodiment of the present invention provides an infrared image data compression method for an infrared imaging simulation system, the method comprising the following steps:

[0006] An infrared imaging simulation system is used to obtain an infrared image, the infrared image is divided into image blocks, a DCT coefficient matrix of the image block is obtained, and the high-frequency information significance of the image block is determined according to the values ​​and position distribution of the coefficients contained in the DCT coefficient matrix of the image block, and the high-frequency area in the image block is determined according to the high-frequency information significance of all the image blocks divided from the infrared image;

[0007] Determine the number of coefficients to be retained in the DCT coefficient matrix of the high-frequency region according to the differences between the coefficients contained in the DCT coefficient matrix of the high-frequency region and the differences between the high-frequency information significance of all image blocks in the infrared image where the high-frequency region is located, and determine the retained coefficients in the DCT coefficient matrix of the high-frequency region according to the number of retained coefficients;

[0008] Using DCT discrete cosine transform, compress the infrared image according to the retention coefficients in the DCT coefficient matrix of all high-frequency areas in the infrared image, perform inverse DCT discrete cosine transform on the compression result to obtain a reconstructed infrared image, record the pixels in the reconstructed infrared image as high-frequency information pixels, determine the trajectory area according to the distance between the high-frequency information pixels in the reconstructed infrared image, and determine the trajectory retention degree of the trajectory area according to the DCT coefficient matrix of the high-frequency area where the corresponding pixel points in the infrared image of the high-frequency information pixel points contained in the trajectory area are located, and the distance between the high-frequency information pixel points contained in different trajectory areas;

[0009] Iterate the coefficient retention quantity with a preset step size to obtain the update trajectory retention degree corresponding to each iterative value, determine the retention quantity suitability of the coefficient retention quantity according to the difference between the update trajectory retention degrees of the coefficient retention quantities of adjacent values ​​and the difference between the trajectory retention degrees of different trajectory areas, and determine the data compression result of the infrared image according to all the retention quantity suitabilities;

[0010] Furthermore, the determining of the saliency of the high-frequency information of the image block includes the following specific methods:

[0011] The coefficient in the lower left corner and the coefficient in the upper right corner of the DCT coefficient matrix of the image block are connected by a line segment, and the cumulative sum of all coefficients in the DCT coefficient matrix within the lower right side of the line segment is recorded as the sum of high-frequency information coefficients of the DCT coefficient matrix of the image block; the maximum value of the sum of high-frequency information coefficients of the DCT coefficient matrix of all image blocks divided from the infrared image where the image block is located is recorded as the maximum high-frequency information coefficient of the image block;

[0012] The ratio of the sum of the high-frequency information coefficients of the DCT coefficient matrix of the image block to the maximum high-frequency information coefficient of the image block is recorded as the high-frequency information ratio of the image block;

[0013] The distance between the maximum value of the coefficients contained in the DCT coefficient matrix of the image block and the coefficient in the upper right corner of the DCT coefficient matrix is ​​recorded as the high-frequency information distance of the DCT coefficient matrix of the image block, and the ratio of the high-frequency information ratio of the image block to the high-frequency information distance of the DCT coefficient matrix of the image block is recorded as the high-frequency information significance of the image block.

[0014] Furthermore, the determining of the high-frequency area in the image block includes the following specific methods:

[0015] When the normalized value of the high-frequency information significance of an image block is greater than a preset high-frequency region screening threshold, the image block is recorded as a high-frequency region.

[0016] Further, the method of determining the number of coefficients to be retained in the DCT coefficient matrix of the high-frequency region includes:

[0017] The range of the coefficients contained in the DCT coefficient matrix of the high-frequency region is recorded as the maximum difference of the coefficients of the DCT coefficient matrix of the high-frequency region, the maximum value of the high-frequency information significance of all image blocks in the infrared image where the high-frequency region is located is recorded as the maximum high-frequency information significance of the high-frequency region, and the ratio of the high-frequency information significance of the high-frequency region to the maximum high-frequency information significance is recorded as the high-frequency information significance ratio of the high-frequency region;

[0018] The normalized value of the ratio of the significant ratio of the high-frequency information in the high-frequency area to the maximum difference in the coefficients of the DCT coefficient matrix in the high-frequency area is recorded as the retention coefficient of the DCT coefficient matrix in the high-frequency area, and the rounded value of the product of the number of coefficients contained in the DCT coefficient matrix in the high-frequency area and the retention coefficient of the DCT coefficient matrix in the high-frequency area is recorded as the number of coefficients retained in the DCT coefficient matrix in the high-frequency area.

[0019] Further, the method of determining the reserved coefficients in the DCT coefficient matrix of the high-frequency region according to the number of reserved coefficients includes the following specific methods:

[0020] The largest number of coefficients among the coefficients contained in the DCT coefficient matrix of the high-frequency region is retained and recorded as the retained coefficients.

[0021] Further, the specific method of determining the trajectory area includes:

[0022] The maximum value of the Euclidean distance between high-frequency information pixels in the reconstructed infrared image is recorded as the maximum reconstruction distance, and the ratio of the Euclidean distance between two high-frequency information pixels in the reconstructed infrared image to the maximum reconstruction distance is recorded as the reconstruction distance ratio of the two high-frequency information pixels;

[0023] When the reconstruction distance ratio of two high-frequency information pixels is less than the reconstruction distance threshold, the two high-frequency information pixels are divided into the same group, and the area composed of all high-frequency information pixels in the same group is recorded as the trajectory area.

[0024] Further, the track retention degree of the track area is determined according to the DCT coefficient matrix of the high-frequency area where the pixel points corresponding to the high-frequency information pixel points contained in the track area in the infrared image are located, and the distances between the high-frequency information pixel points contained in different track areas, including the specific method of:

[0025] The pixel points corresponding to the high-frequency information pixel points contained in the trajectory area in the infrared image are recorded as the original corresponding points of the high-frequency information pixel points, and the mean value of all the retained coefficients in the DCT coefficient matrix of the high-frequency area where the original corresponding points are located is recorded as the frequency of the high-frequency information pixel points;

[0026] Any trajectory area contained in the reconstructed infrared image is recorded as the target trajectory area. The trajectory retention degree of the trajectory area is determined according to the trajectory distance between the target trajectory area and all other trajectory areas contained in the reconstructed infrared image, and the frequency of all high-frequency information pixels contained in the target trajectory area.

[0027] Further, the track retention degree of the track area is determined according to the track distance between the target track area and all other track areas contained in the reconstructed infrared image, the frequency of all high-frequency information pixels contained in the target track area, and the track distance between the track area and all other track areas contained in the reconstructed infrared image where the track area is located, including the specific method of:

[0028] The average frequency of all high-frequency information pixels contained in the target trajectory area is recorded as the frequency concentration of the target trajectory area;

[0029] The cumulative sum of the track distances between the target track area and all other track areas contained in the reconstructed infrared image is recorded as the track distance sum of the target track area;

[0030] The ratio of the frequency concentration of the target trajectory area to the sum of the trajectory distances is recorded as the trajectory retention degree of the target trajectory area.

[0031] Furthermore, the determination of the suitability of the retention number of the coefficient retention number includes the following specific methods:

[0032] The ratio of the track retention degree of the target track area to the maximum value of the track retention degrees of all track areas included in the reconstructed infrared image is recorded as the first ratio of the target track area;

[0033] The sum of the coefficient retention number of the DCT coefficient matrix in the high-frequency region and the number 1 is recorded as the first updated coefficient retention number of the DCT coefficient matrix in the high-frequency region, and the first updated coefficient retention number in the DCT coefficient matrix in the high-frequency region is used as the coefficient that needs to be retained in the compression process to obtain the first updated trajectory retention degree of the trajectory region;

[0034] The absolute value of the difference between the track retention degree of the track area and the first updated track retention degree of the track area is recorded as the first difference of the track area, and the ratio of the first ratio of the track area to the first difference is recorded as the second ratio of the track area;

[0035] The average of the second ratios of all trajectory areas contained in the reconstructed infrared image is recorded as the retention number suitability of the coefficient retention number.

[0036] Furthermore, the data compression result of the infrared image is determined according to the suitability of all the retained quantities, and the specific method includes:

[0037] The reconstructed infrared image obtained by inverse discrete cosine transform corresponding to the maximum value of all retained quantity fitness is used as the data compression result of the infrared image.

[0038] The beneficial effects of the present invention are:

[0039] Since the purpose of preserving the infrared image acquired by the infrared imaging simulation system is to preserve the motion trajectory of the target, the present application evaluates the amount of high-frequency information corresponding to the motion trajectory of the target in the DCT coefficient matrix of the image block, obtains the significance of the high-frequency information of the determined image block, and determines the high-frequency area in the image block based on the significance of the high-frequency information. When the significance of the high-frequency information in the high-frequency area is greater, the possibility that the image block contains the motion trajectory of the target is greater, the amount of high-frequency information in the high-frequency area is greater, and the frequency range corresponding to the high-frequency information is larger. In order to ensure the clarity of the target trajectory contained in the infrared image data, the number of coefficients retained in the DCT coefficient matrix of the high-frequency area should be more. Therefore, the number of coefficients retained in the DCT coefficient matrix of the high-frequency area is determined according to the high-frequency area, and the retained coefficients in the DCT coefficient matrix of the high-frequency area are determined; further, according to the infrared image, The retained coefficients in the DCT coefficient matrix of all high-frequency areas are used to compress and decompress the infrared image to obtain the reconstructed infrared image. The clarity of the trajectory area is evaluated according to the trajectory area identified in the reconstructed infrared image, and the trajectory retention degree of the trajectory area is obtained. In order to screen out the most appropriate coefficient retention number and the corresponding retention coefficient, the coefficient retention number is increased with the number 1 as the step size, and the updated trajectory retention degree corresponding to each value of the coefficient retention number is determined, and the retention number suitability of each coefficient retention number is determined. The retention number suitability is an evaluation of the clarity of the selected coefficient retention number for the retention of the target trajectory. Finally, the data compression result of the infrared image is determined according to the suitability of all retention numbers, so as to solve the problem that the target trajectory information contained in the compressed infrared image data is unclear due to inappropriate frequency selection, and improve the accuracy of the information in the compressed infrared image data. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0041] Figure 1 A schematic flow chart of an infrared image data compression method for an infrared imaging simulation system provided by an embodiment of the present invention;

[0042] Figure 2A high-frequency region acquisition flow chart provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0043] 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 only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 , which shows a flow chart of an infrared image data compression method for an infrared imaging simulation system provided by an embodiment of the present invention, the method comprising the following steps:

[0045] Step S001, using an infrared imaging simulation system to acquire an infrared image, dividing the infrared image into image blocks, acquiring a DCT coefficient matrix of the image block, determining the high-frequency information significance of the image block according to the values ​​and position distribution of the coefficients contained in the DCT coefficient matrix of the image block, and determining the high-frequency area in the image block according to the high-frequency information significance of all image blocks divided from the infrared image.

[0046] An infrared imaging simulation system is used to acquire a first preset number of infrared image frames within one second, and the infrared image is divided into image blocks of equal size.

[0047] Preferably, as an embodiment of the present application, the value of the first preset number in this embodiment is 300, and the side length of each image block is 16, that is, each image block is composed of 16 16 pixels are arranged.

[0048] When the edge of the infrared image cannot be divided into a complete image block, the number 0 is used to supplement the incomplete image block to ensure that the size of all image blocks is consistent.

[0049] Perform DCT discrete cosine transform on each image block to obtain the DCT coefficient matrix of the image block.

[0050] It can be understood that the size of the DCT coefficient matrix is ​​the same as the corresponding image block, each position in the DCT coefficient matrix corresponds to a coefficient, the coefficient in the upper left corner of the DCT coefficient matrix corresponds to the low-frequency information of the image block, the coefficient in the lower right corner corresponds to the high-frequency information of the image block, the upper right corner corresponds to the vertical detail information of the image block, and the lower left corner corresponds to the horizontal detail information of the image block. It should be noted that a large amount of noise information is contained between the low-frequency information and the high-frequency information in the DCT coefficient matrix.

[0051] The motion trajectory of the target in the infrared image is generally continuous and smooth, and the target is small and moves fast. Therefore, the motion trajectory of the target often corresponds to the high-frequency information in the image block. In order to ensure the accuracy of the motion trajectory of the target in the infrared image, it is necessary to accurately screen the high-frequency information in the image block.

[0052] When the coefficient of the position of the high-frequency information in the image block in the DCT coefficient matrix of the image block is larger, the high-frequency information contained in the image block is more obvious, and the possibility that the image block contains the motion trajectory of the target is greater.

[0053] The coefficient at the lower left corner and the coefficient at the upper right corner of the DCT coefficient matrix of the image block are connected by a line segment, and the cumulative sum of all coefficients in the DCT coefficient matrix within the lower right range of the line segment is recorded as the sum of high-frequency information coefficients of the DCT coefficient matrix of the image block.

[0054] When the sum of the high-frequency information coefficients of the DCT coefficient matrix of the image block is larger, the coefficient of the high-frequency information in the DCT coefficient matrix of the image block corresponding to the image block is larger, the high-frequency information contained in the image block is more obvious, and the possibility that the image block contains the motion trajectory of the target is greater.

[0055] The distance between the maximum value of the coefficients included in the DCT coefficient matrix of the image block and the coefficient in the upper right corner of the DCT coefficient matrix is ​​recorded as the high-frequency information distance of the DCT coefficient matrix of the image block.

[0056] When the high-frequency information distance of the DCT coefficient matrix of the image block is smaller, the position of the high-frequency information in the corresponding image block in the DCT coefficient matrix of the image block is closer to the position of the maximum value of the coefficient contained in the DCT coefficient matrix, the more obvious the high-frequency information contained in the image block is, and the greater the possibility that the image block contains the motion trajectory of the target.

[0057] The high-frequency information significance of the image block is determined according to the sum of high-frequency information coefficients of the DCT coefficient matrices of all image blocks divided from the infrared image where the image block is located and the high-frequency information distance of the DCT coefficient matrix of the image block.

[0058] Preferably, as an embodiment of the present application, the maximum value of the sum of the high-frequency information coefficients of the DCT coefficient matrices of all image blocks divided from the infrared image where the image block is located is recorded as the maximum high-frequency information coefficient of the image block, and the ratio of the sum of the high-frequency information coefficients of the DCT coefficient matrix of the image block to the maximum high-frequency information coefficient of the image block is recorded as the high-frequency information ratio of the image block. The ratio of the high-frequency information ratio of the image block to the high-frequency information distance of the DCT coefficient matrix of the image block is recorded as the high-frequency information saliency of the image block.

[0059] It should be noted that in the process of calculating the high-frequency information significance of the image block, the high-frequency information distance of the DCT coefficient matrix of the image block is used as the denominator of the fraction corresponding to the ratio. In order to avoid the situation where the denominator is zero, a preset value needs to be added to the denominator. The preset value is 1 in an implementation example.

[0060] When the sum of the high-frequency information coefficients of the DCT coefficient matrix of the image block is larger and the high-frequency information distance of the DCT coefficient matrix of the image block is smaller, the high-frequency information contained in the image block is more obvious and the possibility that the image block contains the motion trajectory of the target is greater. At this time, the high-frequency information significance of the image block is greater.

[0061] The normalized value of the high-frequency information saliency of the image block is compared with the high-frequency region screening threshold. When the normalized value of the high-frequency information saliency of the image block is greater than the high-frequency region screening threshold, the image block is recorded as a high-frequency region.

[0062] In this embodiment, the high-frequency region screening threshold is set to 0.7.

[0063] It should be noted that this embodiment uses the Z-Score standard normalization method to calculate the normalized value. In actual application, the implementer may use other methods in the prior art such as the maximum and minimum value normalization method, the sigmoid function, etc. to calculate the normalized value, which is not limited here.

[0064] At this point, all high-frequency areas in each frame of infrared image are obtained. The high-frequency area acquisition flow chart is as follows: Figure 2 shown.

[0065] Step S002: determine the number of coefficients to be retained in the DCT coefficient matrix of the high-frequency region based on the differences between the coefficients contained in the DCT coefficient matrix of the high-frequency region and the differences between the high-frequency information significance of all image blocks in the infrared image where the high-frequency region is located, and determine the retained coefficients in the DCT coefficient matrix of the high-frequency region based on the number of retained coefficients.

[0066] When the significance of high-frequency information in the high-frequency area is greater, the possibility that the image block contains the motion trajectory of the target is greater, the amount of high-frequency information in the high-frequency area is more, and the frequency range corresponding to the high-frequency information is larger. In order to ensure the clarity of the target trajectory contained in the infrared image data, the more coefficients in the DCT coefficient matrix of the high-frequency area should be retained, and further, the frequency set in the compression process determined by the retained coefficients in the DCT coefficient matrix is ​​more accurate.

[0067] The number of coefficients retained in the DCT coefficient matrix of the high-frequency region is determined based on the differences between the coefficients contained in the DCT coefficient matrix of the high-frequency region and the differences between the high-frequency information significances of all image blocks in the infrared image where the high-frequency region is located.

[0068] Preferably, as an embodiment of the present application, the range of the coefficients contained in the DCT coefficient matrix of the high-frequency region is recorded as the maximum difference in the coefficients of the DCT coefficient matrix of the high-frequency region, the maximum value of the high-frequency information significance of all image blocks in the infrared image where the high-frequency region is located is recorded as the maximum high-frequency information significance of the high-frequency region, the ratio of the high-frequency information significance of the high-frequency region to the maximum high-frequency information significance is recorded as the high-frequency information significance ratio of the high-frequency region, the normalized value of the ratio of the high-frequency information significance ratio of the high-frequency region to the maximum difference in the coefficients of the DCT coefficient matrix of the high-frequency region is recorded as the retention coefficient of the DCT coefficient matrix of the high-frequency region, and the rounded value of the product of the number of coefficients contained in the DCT coefficient matrix of the high-frequency region and the retention coefficient of the DCT coefficient matrix of the high-frequency region is recorded as the number of retained coefficients of the DCT coefficient matrix of the high-frequency region.

[0069] It should be noted that this embodiment uses the Z-Score standard normalization method to calculate the normalized value. In actual application, the implementer may use other methods in the prior art such as the maximum and minimum value normalization method, the sigmoid function, etc. to calculate the normalized value, which is not limited here.

[0070] The number of coefficients retained in the DCT coefficient matrix of the high-frequency region is the number of coefficients selected when retaining the coefficients in the DCT coefficient matrix of the high-frequency region when compressing the infrared image, so the value of the number of coefficients retained should be an integer greater than or equal to 1 and less than or equal to 256, where 256 is the number of pixels contained in the image block. When the high-frequency information significance of the high-frequency region is greater relative to the maximum high-frequency information significance of the high-frequency region, and the range of the coefficients contained in the DCT coefficient matrix of the high-frequency region is smaller, the amount of high-frequency information contained in the high-frequency region is greater, the frequency range corresponding to the high-frequency information is larger, and the possibility of the target's motion trajectory contained in the high-frequency region is greater. In order to ensure the clarity of the target trajectory contained in the infrared image data, the number of coefficients retained in the DCT coefficient matrix of the high-frequency region should be greater.

[0071] The largest number of coefficients among the coefficients contained in the DCT coefficient matrix of the high-frequency region is retained and recorded as the retained coefficients.

[0072] At this point, the remaining coefficients in the DCT coefficient matrix of the high frequency region are determined.

[0073] Step S003, using DCT discrete cosine transform, compressing the infrared image according to the retention coefficients in the DCT coefficient matrix of all high-frequency areas in the infrared image, performing inverse DCT discrete cosine transform on the compression result to obtain a reconstructed infrared image, recording the pixels in the reconstructed infrared image as high-frequency information pixels, determining the trajectory area according to the distance between the high-frequency information pixels in the reconstructed infrared image, determining the trajectory retention degree of the trajectory area according to the DCT coefficient matrix of the high-frequency area where the high-frequency information pixels contained in the trajectory area correspond to the pixels in the infrared image, and the distance between the high-frequency information pixels contained in different trajectory areas.

[0074] The mapping relationship of DCT discrete cosine transform is a one-to-one relationship. It can be understood that when DCT discrete cosine transform is used to compress infrared images, the compressed infrared images can be restored to the original infrared images through the inverse transform of DCT discrete cosine transform. In order to make the frequency range selected by the compression process better present the motion trajectory of the target, it is necessary to evaluate the rationality of the number of retained coefficients in the DCT coefficient matrix in the high-frequency area.

[0075] The infrared image is compressed using DCT discrete cosine transform, and the retained coefficients in the DCT coefficient matrix in the high-frequency area are used as the coefficients that need to be retained during the compression process to obtain the compressed data of the infrared image. The compressed data of the infrared image is subjected to inverse discrete cosine transform to obtain the reconstructed infrared image. Since the reconstructed infrared image is determined by compression and decompression based on the high-frequency information in the infrared image, the pixels in the reconstructed infrared image are recorded as high-frequency information pixels.

[0076] The Euclidean distance between every two high-frequency information pixels in the reconstructed infrared image is calculated, and the maximum value of the Euclidean distance between the high-frequency information pixels in the reconstructed infrared image is recorded as the maximum reconstruction distance. The ratio of the Euclidean distance between two high-frequency information pixels in the reconstructed infrared image to the maximum reconstruction distance is recorded as the reconstruction distance ratio of the two high-frequency information pixels.

[0077] When the reconstruction distance ratio of two high-frequency information pixels is smaller, the two high-frequency information pixels are more likely to correspond to close positions in the target's motion trajectory.

[0078] The reconstruction distance ratio of the two high-frequency information pixels is compared with the reconstruction distance threshold. When the reconstruction distance ratio of the two high-frequency information pixels is less than the reconstruction distance threshold, the two high-frequency information pixels are divided into the same group. The area composed of all high-frequency information pixels in the same group is recorded as the trajectory area. In this embodiment, the value of the reconstruction distance threshold is 0.4.

[0079] The minimum value of the Euclidean distance between the high-frequency information pixels contained in the two trajectory areas is recorded as the trajectory distance of the two trajectory areas.

[0080] It should be noted that when the reconstructed infrared image can only be divided into one trajectory area, the trajectory distance between the two trajectory areas cannot be calculated. In this case, the trajectory distance between the trajectory area and all other trajectory areas is directly recorded as 0.

[0081] The pixel points corresponding to the high-frequency information pixels contained in the trajectory area in the infrared image are recorded as the original corresponding points of the high-frequency information pixels, and the mean of all retained coefficients in the DCT coefficient matrix of the high-frequency area where the original corresponding points are located is recorded as the frequency of the high-frequency information pixels.

[0082] When the frequency of high-frequency information pixels is larger, the selection of retention coefficients in the DCT coefficient matrix of the high-frequency area is more reasonable. The retention coefficients in the DCT coefficient matrix of the high-frequency area are used as the coefficients that need to be retained in the compression process. The compressed data of the acquired infrared image can more clearly retain the motion trajectory of the target.

[0083] Any trajectory area contained in the reconstructed infrared image is recorded as the target trajectory area. The trajectory retention degree of the trajectory area is determined according to the trajectory distance between the target trajectory area and all other trajectory areas contained in the reconstructed infrared image, and the frequency of all high-frequency information pixels contained in the target trajectory area.

[0084] Preferably, as an embodiment of the present application, the average frequency of all high-frequency information pixel points contained in the target trajectory area is recorded as the frequency concentration of the target trajectory area, the cumulative sum of the trajectory distances between the target trajectory area and all other trajectory areas contained in the reconstructed infrared image is recorded as the trajectory distance sum of the target trajectory area, and the ratio of the frequency concentration of the target trajectory area to the trajectory distance sum is recorded as the trajectory retention degree of the target trajectory area.

[0085] It should be noted that in the process of calculating the trajectory retention degree of the trajectory area, in order to avoid the situation where the denominator of the fraction corresponding to the trajectory distance of the trajectory area and the ratio is zero, a preset value needs to be added to the denominator. The implementation example of the preset value is 1. When the reconstructed infrared image can only divide one trajectory area, the denominator will be zero.

[0086] When the retention coefficients in the DCT coefficient matrix of the high-frequency area are different, the amount of information about the target's motion trajectory contained in the compressed data of the infrared image is different. Specifically, when the frequency of all high-frequency information pixels contained in the trajectory area is greater and the sum of the trajectory distances in the trajectory area is smaller, the selection of the retention coefficients in the DCT coefficient matrix of the high-frequency area is more reasonable, and the target trajectory contained in the compressed data of the infrared image is clearer.

[0087] Step S004, iterate the coefficient retention number with a preset step size, obtain the update trajectory retention degree corresponding to each iterative value, determine the retention number suitability of the coefficient retention number according to the difference between the update trajectory retention degrees of the coefficient retention numbers of adjacent values ​​and the difference between the trajectory retention degrees of different trajectory areas, and determine the data compression result of the infrared image according to all the retention number suitabilities.

[0088] The sum of the coefficient retention number of the DCT coefficient matrix in the high-frequency region and the number 1 is recorded as the first updated coefficient retention number of the DCT coefficient matrix in the high-frequency region. The first updated coefficient retention number in the DCT coefficient matrix in the high-frequency region is used as the coefficient that needs to be retained during the compression process, and the first updated trajectory retention degree of the trajectory region is obtained according to the method of obtaining the trajectory retention degree of the trajectory region based on the retained coefficients in the DCT coefficient matrix in the high-frequency region.

[0089] It can be understood that the first updated trajectory retention degree of the trajectory area is the trajectory retention degree corresponding to the coefficient retention number when the coefficient retention number increases to the first updated coefficient retention number.

[0090] The ratio of the track retention degree of the track area to the maximum value of the track retention degrees of all track areas contained in the reconstructed infrared image where the track area is located is recorded as the first ratio of the track area, the absolute value of the difference between the track retention degree of the track area and the first updated track retention degree of the track area is recorded as the first difference of the track area, and the ratio of the first ratio of the track area to the first difference is recorded as the second ratio of the track area. The average of the second ratios of all track areas contained in the reconstructed infrared image is recorded as the retention quantity suitability of the coefficient retention quantity.

[0091] When the first ratio of the trajectory area is larger, the possibility that the trajectory area corresponds to the motion trajectory of the target point is greater, and the motion trajectory of the target corresponding to the trajectory area is retained more completely and clearly. At the same time, when the trajectory retention degree of the trajectory area and the first updated trajectory retention degree of the trajectory area are smaller, the reconstructed infrared image determined by the coefficient retention number retains the motion trajectory of the target more completely and clearly. At this time, the suitability of the coefficient retention number is greater.

[0092] Continue to repeat the above steps, calculate the sum of the first updated coefficient retention number of the DCT coefficient matrix in the high-frequency region and the number 1, and obtain the second updated coefficient retention number of the DCT coefficient matrix in the high-frequency region. The second updated coefficient retention number in the DCT coefficient matrix in the high-frequency region is used as the coefficient that needs to be retained in the compression process, and the second updated trajectory retention degree of the trajectory region is obtained according to the method of obtaining the trajectory retention degree of the trajectory region according to the retained coefficients in the DCT coefficient matrix in the high-frequency region.

[0093] It can be understood that the second updated trajectory retention degree of the trajectory area is the trajectory retention degree corresponding to the coefficient retention number when the coefficient retention number increases to the second updated coefficient retention number.

[0094] The ratio of the track retention degree of the track area to the maximum value of the track retention degrees of all track areas contained in the reconstructed infrared image where the track area is located is recorded as the third ratio of the track area, the absolute value of the difference between the track retention degree of the track area and the second updated track retention degree of the track area is recorded as the second difference of the track area, and the ratio of the third ratio of the track area to the second difference is recorded as the fourth ratio of the track area. The average of the fourth ratios of all track areas contained in the reconstructed infrared image is recorded as the retention quantity suitability of the first coefficient retention quantity.

[0095] The above steps are repeated until the number of coefficients retained is increased by 1 each time, and the obtained value is equal to the first preset threshold. The first preset threshold is the number of pixels contained in the image block. In this embodiment, the value of the first preset threshold is 256.

[0096] It can be understood that the number of retained coefficients is the number of coefficients selected when retaining coefficients in the DCT coefficient matrix in the high-frequency region when compressing the infrared image, so the maximum value of the number of retained coefficients increasing by 1 is the first preset threshold.

[0097] The reconstructed infrared image obtained by inverse discrete cosine transform corresponding to the maximum value of all retained quantity fitness is used as the data compression result of the infrared image.

[0098] At this point, the data compression of the infrared image of the infrared imaging simulation system is completed.

[0099] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An infrared image data compression method for an infrared imaging simulation system, characterized in that: The method comprises the following steps: An infrared imaging simulation system is used to obtain an infrared image, the infrared image is divided into image blocks, a DCT coefficient matrix of the image block is obtained, and the high-frequency information significance of the image block is determined according to the values ​​and position distribution of the coefficients contained in the DCT coefficient matrix of the image block, and the high-frequency area in the image block is determined according to the high-frequency information significance of all the image blocks divided from the infrared image; Determine the number of coefficients to be retained in the DCT coefficient matrix of the high-frequency region according to the differences between the coefficients contained in the DCT coefficient matrix of the high-frequency region and the differences between the high-frequency information significance of all image blocks in the infrared image where the high-frequency region is located, and determine the retained coefficients in the DCT coefficient matrix of the high-frequency region according to the number of retained coefficients; Using DCT discrete cosine transform, compress the infrared image according to the retention coefficients in the DCT coefficient matrix of all high-frequency areas in the infrared image, perform inverse DCT discrete cosine transform on the compression result to obtain a reconstructed infrared image, record the pixels in the reconstructed infrared image as high-frequency information pixels, determine the trajectory area according to the distance between the high-frequency information pixels in the reconstructed infrared image, and determine the trajectory retention degree of the trajectory area according to the DCT coefficient matrix of the high-frequency area where the corresponding pixel points in the infrared image of the high-frequency information pixel points contained in the trajectory area are located, and the distance between the high-frequency information pixel points contained in different trajectory areas; Iterate the coefficient retention quantity with a preset step size to obtain the update trajectory retention degree corresponding to each iterative value, determine the retention quantity suitability of the coefficient retention quantity according to the difference between the update trajectory retention degrees of the coefficient retention quantities of adjacent values ​​and the difference between the trajectory retention degrees of different trajectory areas, and determine the data compression result of the infrared image according to all the retention quantity suitabilities; The specific method of determining the saliency of high-frequency information of the image block includes: The coefficient in the lower left corner and the coefficient in the upper right corner of the DCT coefficient matrix of the image block are connected by a line segment, and the cumulative sum of all coefficients in the DCT coefficient matrix within the lower right side of the line segment is recorded as the sum of high-frequency information coefficients of the DCT coefficient matrix of the image block; the maximum value of the sum of high-frequency information coefficients of the DCT coefficient matrix of all image blocks divided from the infrared image where the image block is located is recorded as the maximum high-frequency information coefficient of the image block; The ratio of the sum of the high-frequency information coefficients of the DCT coefficient matrix of the image block to the maximum high-frequency information coefficient of the image block is recorded as the high-frequency information ratio of the image block; The distance between the maximum value of the coefficients contained in the DCT coefficient matrix of the image block and the coefficient in the upper right corner of the DCT coefficient matrix is ​​recorded as the high-frequency information distance of the DCT coefficient matrix of the image block, and the ratio of the high-frequency information ratio of the image block to the high-frequency information distance of the DCT coefficient matrix of the image block is recorded as the high-frequency information significance of the image block.

2. The infrared image data compression method for an infrared imaging simulation system according to claim 1, characterized in that: The specific method of determining the high-frequency area in the image block includes: When the normalized value of the high-frequency information significance of an image block is greater than a preset high-frequency region screening threshold, the image block is recorded as a high-frequency region.

3. The infrared image data compression method for an infrared imaging simulation system according to claim 1, characterized in that: The specific method of determining the number of coefficients to be retained in the DCT coefficient matrix of the high-frequency region includes: The range of the coefficients contained in the DCT coefficient matrix of the high-frequency region is recorded as the maximum difference of the coefficients of the DCT coefficient matrix of the high-frequency region, the maximum value of the high-frequency information significance of all image blocks in the infrared image where the high-frequency region is located is recorded as the maximum high-frequency information significance of the high-frequency region, and the ratio of the high-frequency information significance of the high-frequency region to the maximum high-frequency information significance is recorded as the high-frequency information significance ratio of the high-frequency region; The normalized value of the ratio of the significant ratio of the high-frequency information in the high-frequency area to the maximum difference in the coefficients of the DCT coefficient matrix in the high-frequency area is recorded as the retention coefficient of the DCT coefficient matrix in the high-frequency area, and the rounded value of the product of the number of coefficients contained in the DCT coefficient matrix in the high-frequency area and the retention coefficient of the DCT coefficient matrix in the high-frequency area is recorded as the number of coefficients retained in the DCT coefficient matrix in the high-frequency area.

4. The infrared image data compression method for an infrared imaging simulation system according to claim 1, characterized in that: The specific method of determining the reserved coefficients in the DCT coefficient matrix of the high-frequency region according to the number of reserved coefficients includes: The largest number of coefficients among the coefficients contained in the DCT coefficient matrix of the high-frequency region is retained and recorded as the retained coefficients.

5. The infrared image data compression method for an infrared imaging simulation system according to claim 1, characterized in that: The specific method of determining the trajectory area includes: The maximum value of the Euclidean distance between high-frequency information pixels in the reconstructed infrared image is recorded as the maximum reconstruction distance, and the ratio of the Euclidean distance between two high-frequency information pixels in the reconstructed infrared image to the maximum reconstruction distance is recorded as the reconstruction distance ratio of the two high-frequency information pixels; When the reconstruction distance ratio of two high-frequency information pixels is less than the reconstruction distance threshold, the two high-frequency information pixels are divided into the same group, and the area composed of all high-frequency information pixels in the same group is recorded as the trajectory area.

6. The infrared image data compression method for an infrared imaging simulation system according to claim 1, characterized in that: The method of determining the track retention degree of the track area according to the DCT coefficient matrix of the high-frequency area where the pixel points corresponding to the high-frequency information pixel points contained in the track area in the infrared image are located and the distances between the high-frequency information pixel points contained in different track areas includes: The pixel points corresponding to the high-frequency information pixel points contained in the trajectory area in the infrared image are recorded as the original corresponding points of the high-frequency information pixel points, and the mean value of all the retained coefficients in the DCT coefficient matrix of the high-frequency area where the original corresponding points are located is recorded as the frequency of the high-frequency information pixel points; Any trajectory area contained in the reconstructed infrared image is recorded as the target trajectory area. The trajectory retention degree of the trajectory area is determined according to the trajectory distance between the target trajectory area and all other trajectory areas contained in the reconstructed infrared image, and the frequency of all high-frequency information pixels contained in the target trajectory area.

7. The infrared image data compression method for an infrared imaging simulation system according to claim 6, characterized in that: The method of determining the track retention degree of the track area according to the track distance between the target track area and all other track areas contained in the reconstructed infrared image, the frequency of all high-frequency information pixels contained in the target track area, and the track distance between the track area and all other track areas contained in the reconstructed infrared image where the track area is located comprises the following specific methods: The average frequency of all high-frequency information pixels contained in the target trajectory area is recorded as the frequency concentration of the target trajectory area; The cumulative sum of the track distances between the target track area and all other track areas contained in the reconstructed infrared image is recorded as the track distance sum of the target track area; The ratio of the frequency concentration of the target trajectory area to the sum of the trajectory distances is recorded as the trajectory retention degree of the target trajectory area.

8. The infrared image data compression method for an infrared imaging simulation system according to claim 7, characterized in that: The determination coefficient retention quantity suitability includes the following specific methods: The ratio of the track retention degree of the target track area to the maximum value of the track retention degrees of all track areas included in the reconstructed infrared image is recorded as the first ratio of the target track area; The sum of the coefficient retention number of the DCT coefficient matrix in the high-frequency region and the number 1 is recorded as the first updated coefficient retention number of the DCT coefficient matrix in the high-frequency region, and the first updated coefficient retention number in the DCT coefficient matrix in the high-frequency region is used as the coefficient that needs to be retained in the compression process to obtain the first updated trajectory retention degree of the trajectory region; The absolute value of the difference between the track retention degree of the track area and the first updated track retention degree of the track area is recorded as the first difference of the track area, and the ratio of the first ratio of the track area to the first difference is recorded as the second ratio of the track area; The average of the second ratios of all trajectory areas contained in the reconstructed infrared image is recorded as the retention number suitability of the coefficient retention number.

9. The infrared image data compression method for an infrared imaging simulation system according to claim 1, characterized in that: The specific method of determining the data compression result of the infrared image according to the suitability of all the retained quantities is as follows: The reconstructed infrared image obtained by inverse discrete cosine transform corresponding to the maximum value of all retained quantity fitness is used as the data compression result of the infrared image.

Images