Multi-channel ISP processing method, device, equipment and medium for infrared images
By denoising and dynamic range compression of infrared images, and using gamma correction and Sobel filtering technology to generate multiple style images, the problem of single dynamic range of infrared images is solved, and diversified contrast display effects and efficient processing are achieved.
Patent Information
- Application Number
- CN202410414380.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-08
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-04-08
AI Technical Summary
The single 14-bit dynamic range of infrared images cannot meet the diverse contrast requirements, resulting in the inability to obtain multiple image styles of infrared images and cannot meet the user's display effect requirements.
By denoising the original infrared image, the detailed image is obtained and dynamic range compression is performed, the image superposition and filtering is performed based on different preset coefficients and directions to generate multiple styles of targets Images and stitch them in the display interface.
It realizes multiple infrared image displays in different styles, improves image quality and display effects, reduces redundant threads of computing resources, and improves infrared image processing efficiency.
Smart Images

Figure CN118134786B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of image processing technology, and in particular to a multi-channel ISP processing method, device, equipment and medium for infrared images. Background Art
[0002] With the development of infrared technology, the cost is gradually reduced, and various types of infrared thermal imaging equipment have entered the consumer field, such as mobile phone plug-ins, night vision binoculars, etc. Visible light usually has three channels of RGB (Red, Green, Blue), each channel has 8bit (bit) dynamic range of color information, and the picture is very rich. However, infrared thermal imagers usually have only one channel with a 14bit dynamic range, so the image needs to be displayed on an 8bit grayscale image after dynamic range compression. However, the contrast display effect of each area cannot be guaranteed on an infrared 8bit image, that is, one style of infrared image cannot meet the use requirements. For example, when the contrast of the thermal target needs to be improved, the background contrast needs to be lowered. When the background contrast needs to be improved, the contrast of the thermal target needs to be lowered.
[0003] From the above, it can be seen that how to obtain infrared images of multiple image styles is a problem to be solved in this field. Summary of the invention
[0004] In view of this, the purpose of the present invention is to provide a multi-channel ISP processing method, device, equipment and medium for infrared images, which can obtain infrared images of multiple image styles. The specific scheme is as follows:
[0005] In a first aspect, the present application discloses a multi-channel ISP processing method for infrared images, comprising:
[0006] De-noising the original infrared image to obtain the target denoised image;
[0007] Acquire a detail image of the target denoised image, and perform dynamic range compression on the target denoised image to obtain a first compressed image;
[0008] Performing dynamic range compression on the first compressed image to obtain a second compressed image, and mapping the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image;
[0009] superimposing the detail image onto the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image respectively, and superimposing the detail image onto the processed image to obtain a third style target image;
[0010] performing Sobel filtering in different directions on the second compressed image to obtain filtered images in different directions, and superimposing the filtered images to obtain a fourth style target image;
[0011] The first style target image, the second style target image, the third style target image, and the fourth style target image are displayed in a display interface.
[0012] Optionally, the denoising process is performed on the original infrared image to obtain a target denoised image, including:
[0013] Performing non-uniformity correction on the original infrared image to obtain a corrected image;
[0014] Performing time-domain filtering on the corrected image using an infinite impulse response filter to eliminate the time-domain noise of the corrected image and obtain a time-domain denoised image;
[0015] A bilateral filtering algorithm is used to remove spatial noise from the temporal denoised image to obtain a target denoised image.
[0016] Optionally, obtaining a detail image of the target denoised image includes:
[0017] A convolution operation is performed on the target denoised image using a Laplacian sharpening operator to obtain a detail image of the target denoised image.
[0018] Optionally, the first compressed image is an image of a first preset bit, and the second compressed image and the processed image are both images of a second preset bit; wherein the first preset bit is greater than the second preset bit.
[0019] Optionally, performing Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposing the filtered images to obtain a fourth style target image includes:
[0020] Performing Sobel filtering in the X direction and Sobel filtering in the Y direction on the second compressed image, respectively, to obtain a first filtered image in the X direction and a second filtered image in the Y direction, respectively;
[0021] The absolute value of the first filtered image and the absolute value of the second filtered image are superimposed to obtain a fourth style target image.
[0022] Optionally, displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in a display interface includes:
[0023] Displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in a spliced form in a display interface;
[0024] When an image zoom instruction is received, a first target image corresponding to the image zoom instruction is screened out from the first style target image, the second style target image, the third style target image, and the fourth style target image, and image zoom processing is performed on the first target image to obtain a zoomed image, so as to display the zoomed image in the display interface.
[0025] Optionally, the process of displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in the display interface further includes:
[0026] When an image movement instruction is received, a second target image corresponding to the image movement instruction is selected from the first style target image, the second style target image, the third style target image, and the fourth style target image, and the second target image is moved.
[0027] In a second aspect, the present application discloses a multi-channel ISP processing device for infrared images, comprising:
[0028] An image noise removal module is used to perform denoising on the original infrared image to obtain a target denoised image;
[0029] A first image compression module, used for acquiring a detail image of the target denoised image, and performing dynamic range compression on the target denoised image to obtain a first compressed image;
[0030] a second image compression module, configured to perform dynamic range compression on the first compressed image to obtain a second compressed image, and map the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image;
[0031] a first image superposition module, configured to superimpose the detail image onto the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image, respectively, and to superimpose the detail image onto the processed image to obtain a third style target image;
[0032] a second image superposition module, configured to perform Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimpose the filtered images to obtain a fourth style target image;
[0033] The target image display module is used to display the first style target image, the second style target image, the third style target image and the fourth style target image in a display interface.
[0034] In a third aspect, the present application discloses an electronic device, comprising:
[0035] Memory, used to store computer programs;
[0036] The processor is used to execute the computer program to implement the steps of the multi-channel ISP processing method of infrared images disclosed above.
[0037] In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the steps of the multi-channel ISP processing method of infrared images disclosed above are implemented.
[0038] The beneficial effects of the present application are as follows: the present application performs denoising processing on the original infrared image to obtain a target denoised image; obtains a detail image of the target denoised image, and performs dynamic range compression on the target denoised image to obtain a first compressed image; performs dynamic range compression on the first compressed image to obtain a second compressed image, and maps the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image; superimposes the detail image on the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image, respectively, and superimposes the detail image on the processed image to obtain a third style target image; performs Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposes the filtered images to obtain a fourth style target image; and displays the first style target image, the second style target image, the third style target image, and the fourth style target image in a display interface. It can be seen that the present application performs denoising on the original infrared image to improve the image quality, obtains a detail image of the target denoised image, performs dynamic range compression on the detail image twice to obtain a first compressed image and a second compressed image respectively, and then uses a gamma correction algorithm based on a lookup table method to map the first compressed image to obtain a processed image, and then superimposes the detail image on the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image respectively, and superimposes the detail image on the processed image to obtain a third style target image, performs Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposes the filtered images to obtain a fourth style target image, such as In this way, on the one hand, the present application can acquire and display multiple first style target images, second style target images, third style target images and fourth style target images of different styles in the display interface, realize multi-channel ISP processing of infrared images, improve the display effect, and allow users to use target images of different styles for multi-dimensional observation. On the other hand, the target denoised image, detail image, first compressed image, second compressed image and processed image in the present application can be used as image resources required to acquire target images of different styles. That is to say, the algorithm threads for acquiring multi-style target images are integrated to reduce redundant threads. Compared with the prior art, multiple independent algorithm threads are required to process when acquiring multiple style infrared images, which can significantly reduce computing resources and improve infrared image processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the embodiments of the present application 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 embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0040] Figure 1 A flowchart of a multi-channel ISP processing method for infrared images disclosed in this application;
[0041] Figure 2 A specific gamma curve schematic diagram disclosed in this application;
[0042] Figure 3 A schematic diagram of a multi-channel ISP processing thread for a specific infrared image disclosed in this application;
[0043] Figure 4 A specific schematic diagram of displaying target images of multiple styles disclosed in this application;
[0044] Figure 5 This is a first specific thumbnail display schematic diagram disclosed in this application;
[0045] Figure 6 This is a second specific thumbnail display schematic diagram disclosed in this application;
[0046] Figure 7 This is a schematic diagram of the structure of a multi-channel ISP processing device for infrared images disclosed in this application;
[0047] Figure 8 This is a structural diagram of an electronic device disclosed in this application. DETAILED DESCRIPTION
[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. 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 making creative work are within the scope of protection of the present invention.
[0049] With the development of infrared technology, the cost is gradually reduced, and various types of infrared thermal imaging equipment have entered the consumer field, such as mobile phone plug-ins, night vision binoculars, etc. Visible light usually has three channels of RGB, each channel has 8-bit dynamic range of color information, and the picture is very rich. However, infrared thermal imagers usually have only one channel with a 14-bit dynamic range, so the image needs to be displayed on an 8-bit grayscale image after dynamic range compression. However, the contrast display effect of each area cannot be guaranteed on an infrared 8-bit image, that is, one style of infrared image cannot meet the use requirements. For example, when the contrast of the thermal target needs to be improved, the background contrast needs to be lowered. When the background contrast needs to be improved, the contrast of the thermal target needs to be lowered.
[0050] To this end, the present application correspondingly provides a multi-channel ISP processing solution for infrared images, which can obtain infrared images of multiple image styles.
[0051] See also Figure 1 As shown, the embodiment of the present application discloses a multi-channel ISP processing method for infrared images, including:
[0052] Step S11: performing denoising processing on the original infrared image to obtain a target denoised image.
[0053] In this embodiment, the denoising process of the original infrared image to obtain the target denoised image includes: performing non-uniformity correction on the original infrared image to obtain a corrected image; performing time domain filtering on the corrected image using an infinite impulse response filter to eliminate the time domain noise of the corrected image to obtain a time domain denoised image; and using a bilateral filtering algorithm to remove the spatial domain noise of the time domain denoised image to obtain the target denoised image. The process of obtaining the target denoised image is as follows:
[0054] 1) Obtain the original infrared image, perform non-uniformity correction on the original infrared image to obtain a corrected image. The two-point correction formula can be used to eliminate the output response rate difference between pixels and fixed noise. The specific formula is as follows:
[0055] f1(r,c)=(f0(r,c)-P(r,c))×K(r,c)+B_avrg;
[0056] Where (r, c) is the coordinate of the pixel, P(r, c) is the response value matrix to the corrected uniform surface (usually the shutter), representing the fixed noise, K(r, c) is the response rate matrix obtained by two-point calibration, B_avrg is the mean of the response value matrix, f0(r, c) is the input original infrared image, and f1(r, c) is the image after non-uniformity correction, that is, the corrected image;
[0057] 2) Use an infinite-duration impulse response filter (IIR) to perform time-domain filtering on the corrected image to eliminate the time-domain noise of the corrected image and obtain a time-domain denoised image. The specific formula is as follows:
[0058] f2(r,c)=f1(r,c)×ratio+f_last(r,c);
[0059] Where ratio is the coefficient of time domain filtering, f_last(r, c) is the filtered image of the previous frame, and f2(r, c) is the image after time domain filtering, that is, the image after time domain denoising;
[0060] 3) Use bilateral filtering algorithm to remove spatial noise of the image after time domain denoising to obtain the target denoised image; first, determine the spatial domain weight: for each pixel in the denoised image, determine the spatial domain weight w_s of each pixel in the adjacent 5×5 neighborhood. It should be noted that the closer the neighborhood pixel is to the center point, the larger the weight is. The weight can be adjusted according to the required degree of smoothness; secondly, determine the intensity domain weight: determine the intensity domain weight w_r according to the intensity difference between each pixel in the adjacent 5×5 neighborhood and the central pixel. The smaller the difference is, the larger the weight is. The weight can be adjusted according to the required degree of smoothness; finally, calculate the weighted average: calculate the final weight by the spatial domain weight and the intensity domain weight, and perform weighted average on each pixel in the field to obtain the filtered pixel value. The specific formula is as follows:
[0061]
[0062] Where (i, j) is the pixel coordinate of the 5×5 area of the pixel point f2(r, c), w_s(i, j, r, c) is the spatial domain weight, w_r(i, j, r, c) is the intensity domain weight, and f3(r, c) is the image after time domain filtering, that is, the target denoised image.
[0063] It should be noted that in the process of acquiring the target denoised image, image enhancement algorithms and other image denoising methods may also be included to improve the quality and display effect of the target denoised image. After acquiring the target denoised image, the multi-channel ISP processing (image signal processing) of this embodiment is based on the target denoised image, that is, only one target denoised image of the original infrared image needs to be acquired, and subsequent acquisitions of target images of various image styles are based on this target denoised image, and there is no need for each image style target image to correspond to an independent target denoised image, which can reduce redundant threads.
[0064] Step S12: Acquire a detail image of the target denoised image, and perform dynamic range compression on the target denoised image to obtain a first compressed image.
[0065] In this embodiment, the step of obtaining the detail image of the target denoised image includes: performing a convolution operation on the target denoised image using a Laplace sharpening operator to obtain the detail image of the target denoised image. The formula for extracting the detail image is as follows:
[0066] detail(r,c)=f3(r,c)×Laplacian;
[0067] Among them, detail(r, c) represents the detail image, Laplacian represents the Laplacian sharpening operator, and the Laplacian sharpening operator can be represented by the following matrix:
[0068]
[0069] It can be understood that the target denoised image f3 (r, c) is usually 14-bit dynamic range data, and then a common histogram equalization algorithm is used to perform dynamic range image compression on the target denoised image to obtain the first compressed image f4 (r, c).
[0070] In this embodiment, the first compressed image is an image of a first preset bit, wherein the first preset bit is, for example, 10 bits, that is, the first compressed image is a 10-bit image.
[0071] Step S13: Dynamic range compression is performed on the first compressed image to obtain a second compressed image, and the first compressed image is mapped using a gamma correction algorithm based on a lookup table method to obtain a processed image.
[0072] In this embodiment, the second compressed image and the processed image are both images of a second preset bit; wherein the first preset bit is greater than the second preset bit. The process of obtaining the second compressed image and the processed image is as follows:
[0073] 1) The first compressed image is subjected to dynamic range compression to obtain a second compressed image A(r, c), wherein the second compressed image is an image of a second preset bit, the second preset bit is, for example, 8 bits, and the second compressed image is a white-hot mode image. The specific formula is as follows:
[0074] A(r,c)=f4(r,c) / 4;
[0075] 2) Mapping the first compressed image using a gamma correction algorithm based on a look-up table (LUT) method to obtain a processed image B (r, c), the processed image is an image of a second preset bit, the second preset bit is, for example, 8 bits, and the processed image is a high temperature target highlighting mode image, for example Figure 2 A specific gamma curve schematic diagram is shown, the lookup table can be a gamma curve, and the specific formula is as follows:
[0076] y = 256 × (x / 1024) gamma ;
[0077] B(r,c)=LUT(f4(r,c));
[0078] It is understandable, however, that the lookup table may satisfy the gamma curve characteristics or may be a freely set curve, thereby achieving different degrees of high temperature target highlighting images.
[0079] Step S14: superimposing the detail image onto the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image respectively, and superimposing the detail image onto the processed image to obtain a third style target image.
[0080] The detail image is superimposed on the second compressed image based on different preset coefficients to obtain the first style target image and the second style target image respectively. For example, the extracted detail image is superimposed on the second compressed image with coefficients of 2 and 8 respectively to obtain the first style target image C(r, c) and the second style target image D(r, c), wherein the first style target image is a normal white-hot image and the second style target image is a detail-highlighted image. The specific formula is as follows:
[0081] C(r,c)=A(r,c)+2×detail(r,c);
[0082] D(r,c)=A(r,c)+8×detail(r,c).
[0083] The detail image is superimposed on the processed image to obtain a third style target image. Specifically, the detail image can be superimposed on the processed image by a factor of 2 to obtain a third style target image E(r, c), where the third style target image is a high temperature target highlighted image. The specific formula is as follows:
[0084] E(r,c)=B(r,c)+2×detail(r,c).
[0085] Step S15: performing Sobel filtering in different directions on the second compressed image to obtain filtered images in different directions, and superimposing the filtered images to obtain a fourth style target image.
[0086] In this embodiment, the second compressed image is subjected to Sobel filtering in different directions to obtain filtered images in different directions, and each of the filtered images is superimposed to obtain a fourth style target image, including: the second compressed image is subjected to Sobel filtering in the X direction and Sobel filtering in the Y direction to obtain a first filtered image in the X direction and a second filtered image in the Y direction; the absolute value of the first filtered image is superimposed with the absolute value of the second filtered image to obtain a fourth style target image. The specific formula of the Sobel filter operator in the X direction is as follows:
[0087]
[0088] The specific formula of the Sobel filter operator in the Y direction is as follows:
[0089]
[0090] The second compressed image is subjected to Sobel filtering in the X direction and Sobel filtering in the Y direction to obtain a first filtered image in the X direction and a second filtered image in the Y direction, respectively. Then, the absolute value of the first filtered image and the absolute value of the second filtered image are superimposed to obtain a fourth style target image F(r, c), wherein the fourth style target image is an outline image, and the specific formula is as follows:
[0091] F(r,c)=|A(r,c)×sobel_x|+|A(r,c)×sobel_y|.
[0092] For example Figure 3 A specific schematic diagram of multi-channel ISP processing threads for infrared images is shown. In this embodiment, the threads for obtaining target images of multiple styles are not multiple independent threads, but the redundant threads are merged. Compared with the prior art, when obtaining infrared images of multiple styles, multiple independent algorithm threads are required for processing, which can significantly reduce computing resources.
[0093] Step S16: displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in a display interface.
[0094] In this embodiment, the displaying of the first style target image, the second style target image, the third style target image and the fourth style target image in a display interface includes: displaying the first style target image, the second style target image, the third style target image and the fourth style target image in a spliced form in the display interface; when receiving an image zoom instruction, filtering out a first target image corresponding to the image zoom instruction from the first style target image, the second style target image, the third style target image and the fourth style target image, and performing image zoom processing on the first target image to obtain a zoomed image, so as to display the zoomed image in the display interface.
[0095] For example Figure 4 A specific schematic diagram of displaying multiple style target images is shown, in which the first style target image, the second style target image, the third style target image and the fourth style target image are spliced together front to back and left to right and displayed on the screen, so as to facilitate user viewing.
[0096] Furthermore, the user can enlarge or reduce the four target images according to specific usage requirements. That is, when receiving an image zoom instruction, the first target image corresponding to the image zoom instruction is filtered out from the first style target image, the second style target image, the third style target image and the fourth style target image, and the first target image is zoomed to obtain the zoomed image, so as to display the zoomed image in the display interface.
[0097] For example Figure 5 As shown in the first specific thumbnail display diagram, when the user double-clicks the second style target image, that is, receives an image zoom-out instruction, the second style target image will be determined as the first target image, and then the first target image will be zoomed out and presented in the form of a thumbnail. The remaining first style target image, third style target image and fourth style target image will be displayed side by side on the screen.
[0098] For example Figure 6 As shown in the second specific thumbnail display diagram, when the user double-clicks the third style target image, the third style target image will be determined as the first target image, and then the first target image will be reduced to the side of the screen and presented in the form of a thumbnail. The remaining first style target image and fourth style target image will be displayed side by side on the screen.
[0099] The system sets a default thumbnail display area, and users can also customize the thumbnail display area according to specific usage requirements. For example, when there is one thumbnail in the display interface, the thumbnail can be displayed separately at the bottom of the screen. When there are multiple thumbnails in the display interface, the thumbnails can be displayed side by side in the thumbnail area of the screen, such as the thumbnail area on the right side of the display interface.
[0100] It should be noted that the user can also issue an image zoom instruction, which can be for a target image presented in thumbnail form or a target image not presented in thumbnail form; for example, when the user double-clicks the third style target image presented in thumbnail form, it is determined that the user has issued an image zoom instruction, and the third style target image is therefore zoomed in, which can be to restore the third style target image to its size before reduction, or to enlarge the third style target image according to a user-defined zoom ratio; when the user issues an image zoom instruction for a first style target image not presented in thumbnail form, the first style target image can be enlarged, and the enlarged size can be determined based on the user-defined zoom ratio.
[0101] In this embodiment, the process of displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in the display interface further includes: when receiving an image movement instruction, filtering out a second target image corresponding to the image movement instruction from the first style target image, the second style target image, the third style target image, and the fourth style target image, and moving the second target image. It can be understood that the image movement instruction can be for a thumbnail, or for a target image of the original size, or for a target image after being magnified. For example, in the case of multiple large images being displayed, after the user long presses one of the images, the image will move with the user, and the positions of the two images on the screen can be swapped after dragging the image to the position of another image and releasing it.
[0102] The beneficial effects of the present application are as follows: the present application performs denoising processing on the original infrared image to obtain a target denoised image; obtains a detail image of the target denoised image, and performs dynamic range compression on the target denoised image to obtain a first compressed image; performs dynamic range compression on the first compressed image to obtain a second compressed image, and maps the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image; superimposes the detail image on the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image, respectively, and superimposes the detail image on the processed image to obtain a third style target image; performs Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposes the filtered images to obtain a fourth style target image; and displays the first style target image, the second style target image, the third style target image, and the fourth style target image in a display interface. It can be seen that the present application performs denoising on the original infrared image to improve the image quality, obtains a detail image of the target denoised image, performs dynamic range compression on the detail image twice to obtain a first compressed image and a second compressed image respectively, and then uses a gamma correction algorithm based on a lookup table method to map the first compressed image to obtain a processed image, and then superimposes the detail image on the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image respectively, and superimposes the detail image on the processed image to obtain a third style target image, performs Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposes the filtered images to obtain a fourth style target image, such as In this way, on the one hand, the present application can acquire and display multiple first style target images, second style target images, third style target images and fourth style target images of different styles in the display interface, realize multi-channel ISP processing of infrared images, improve the display effect, and allow users to use target images of different styles for multi-dimensional observation. On the other hand, the target denoised image, detail image, first compressed image, second compressed image and processed image in the present application can be used as image resources required to acquire target images of different styles. That is to say, the algorithm threads for acquiring multi-style target images are integrated to reduce redundant threads. Compared with the prior art, multiple independent algorithm threads are required to process when acquiring multiple style infrared images, which can significantly reduce computing resources and improve infrared image processing efficiency.
[0103] See also Figure 7 As shown, the embodiment of the present application discloses a multi-channel ISP processing device for infrared images, including:
[0104] The image noise removal module 11 is used to perform denoising on the original infrared image to obtain a target denoised image;
[0105] A first image compression module 12 is used to obtain a detail image of the target denoised image and perform dynamic range compression on the target denoised image to obtain a first compressed image;
[0106] A second image compression module 13 is used to perform dynamic range compression on the first compressed image to obtain a second compressed image, and map the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image;
[0107] a first image superposition module 14, configured to superimpose the detail image onto the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image, respectively, and to superimpose the detail image onto the processed image to obtain a third style target image;
[0108] A second image superposition module 15 is used to perform Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and to superimpose the filtered images to obtain a fourth style target image;
[0109] The target image display module 16 is configured to display the first style target image, the second style target image, the third style target image, and the fourth style target image in a display interface.
[0110] The beneficial effects of the present application are as follows: the present application performs denoising processing on the original infrared image to obtain a target denoised image; obtains a detail image of the target denoised image, and performs dynamic range compression on the target denoised image to obtain a first compressed image; performs dynamic range compression on the first compressed image to obtain a second compressed image, and maps the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image; superimposes the detail image on the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image, respectively, and superimposes the detail image on the processed image to obtain a third style target image; performs Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposes the filtered images to obtain a fourth style target image; and displays the first style target image, the second style target image, the third style target image, and the fourth style target image in a display interface. It can be seen that the present application performs denoising on the original infrared image to improve the image quality, obtains a detail image of the target denoised image, performs dynamic range compression on the detail image twice to obtain a first compressed image and a second compressed image respectively, and then uses a gamma correction algorithm based on a lookup table method to map the first compressed image to obtain a processed image, and then superimposes the detail image on the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image respectively, and superimposes the detail image on the processed image to obtain a third style target image, performs Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposes the filtered images to obtain a fourth style target image, such as In this way, on the one hand, the present application can acquire and display multiple first style target images, second style target images, third style target images and fourth style target images of different styles in the display interface, realize multi-channel ISP processing of infrared images, improve the display effect, and allow users to use target images of different styles for multi-dimensional observation. On the other hand, the target denoised image, detail image, first compressed image, second compressed image and processed image in the present application can be used as image resources required to acquire target images of different styles. That is to say, the algorithm threads for acquiring multi-style target images are integrated to reduce redundant threads. Compared with the prior art, multiple independent algorithm threads are required to process when acquiring multiple style infrared images, which can significantly reduce computing resources and improve infrared image processing efficiency.
[0111] Furthermore, an embodiment of the present application also provides an electronic device. Figure 8This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content in the diagram cannot be regarded as any limitation on the scope of use of the present application.
[0112] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Specifically, it may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is used to store a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the multi-channel ISP processing method of infrared images performed by the electronic device disclosed in any of the aforementioned embodiments.
[0113] In this embodiment, the power supply 23 is used to provide working voltage for each hardware device on the electronic device; the communication interface 24 can create a data transmission channel between the electronic device and the external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.
[0114] Among them, the processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 21 can be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 21 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 21 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 21 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.
[0115] In addition, the memory 22, as a carrier for storing resources, can be a read-only memory, a random access memory, a disk or an optical disk, etc. The resources stored thereon include an operating system 221, a computer program 222 and data 223, etc. The storage method can be temporary storage or permanent storage.
[0116] The operating system 221 is used to manage and control the hardware devices and computer programs 222 on the electronic device, so as to realize the operation and processing of the massive data 223 in the memory 22 by the processor 21, which can be Windows, Unix, Linux, etc. In addition to including a computer program that can be used to complete the multi-channel ISP processing method of infrared images performed by the electronic device disclosed in any of the aforementioned embodiments, the computer program 222 can further include a computer program that can be used to complete other specific tasks. In addition to data received by the electronic device and transmitted from an external device, the data 223 can also include data collected by its own input and output interface 25, etc.
[0117] Furthermore, the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the multi-channel ISP processing method of infrared images disclosed above is implemented. For the specific steps of the method, reference may be made to the corresponding contents disclosed in the above embodiments, and no further description will be given here.
[0118] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0119] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. The steps of the method or algorithm described in conjunction with the embodiments disclosed herein can be implemented directly with hardware, a software module executed by a processor, or a combination of the two. The software module can be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable EPROM (Erasable Programmable Read Only Memory), an electrically erasable programmable EEPROM (Electrically Erasable Programmable read only memory), a register, a hard disk, a removable disk, a CD-ROM (CoMP24004067act Disc Read-Only Memory), or any other form of storage medium known in the technical field.
[0120] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.
[0121] The above is a detailed introduction to the multi-channel ISP processing method, device, equipment and medium for infrared images provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
Claims
1. A multi-channel ISP processing method for infrared images, characterized in that: include: De-noising the original infrared image to obtain the target denoised image; Acquire a detail image of the target denoised image, and perform dynamic range compression on the target denoised image to obtain a first compressed image; Performing dynamic range compression on the first compressed image to obtain a second compressed image, and mapping the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image; wherein both the second compressed image and the processed image are images of a second preset bit; superimposing the detail image onto the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image respectively, and superimposing the detail image onto the processed image to obtain a third style target image; performing Sobel filtering in different directions on the second compressed image to obtain filtered images in different directions, and superimposing the filtered images to obtain a fourth style target image; The first style target image, the second style target image, the third style target image, and the fourth style target image are displayed in a display interface.
2. The multi-channel ISP processing method of infrared images according to claim 1, characterized in that: The denoising process is performed on the original infrared image to obtain a target denoised image, including: Performing non-uniformity correction on the original infrared image to obtain a corrected image; Performing time-domain filtering on the corrected image using an infinite impulse response filter to eliminate the time-domain noise of the corrected image and obtain a time-domain denoised image; A bilateral filtering algorithm is used to remove spatial noise from the temporal denoised image to obtain a target denoised image.
3. The multi-channel ISP processing method of infrared images according to claim 1, characterized in that: The step of obtaining a detailed image of the target denoised image comprises: A convolution operation is performed on the target denoised image using a Laplacian sharpening operator to obtain a detail image of the target denoised image.
4. The multi-channel ISP processing method of infrared images according to claim 1, characterized in that: The first compressed image is an image of a first preset bit, wherein the first preset bit is greater than the second preset bit.
5. The multi-channel ISP processing method for infrared images according to any one of claims 1 to 4, characterized in that: The step of performing Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimposing the filtered images to obtain a fourth style target image includes: Performing Sobel filtering in the X direction and Sobel filtering in the Y direction on the second compressed image to obtain a first filtered image in the X direction and a second filtered image in the Y direction respectively; The absolute value of the first filtered image and the absolute value of the second filtered image are superimposed to obtain a fourth style target image.
6. The multi-channel ISP processing method of infrared images according to claim 1, characterized in that: The displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in a display interface includes: Displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in a spliced form in a display interface; When an image zoom instruction is received, a first target image corresponding to the image zoom instruction is screened out from the first style target image, the second style target image, the third style target image, and the fourth style target image, and image zoom processing is performed on the first target image to obtain a zoomed image, so as to display the zoomed image in the display interface.
7. The multi-channel ISP processing method of infrared images according to claim 6, characterized in that: The process of displaying the first style target image, the second style target image, the third style target image, and the fourth style target image in the display interface further includes: When an image movement instruction is received, a second target image corresponding to the image movement instruction is selected from the first style target image, the second style target image, the third style target image, and the fourth style target image, and the second target image is moved.
8. A multi-channel ISP processing device for infrared images, characterized in that: include: An image noise removal module is used to perform denoising on the original infrared image to obtain a target denoised image; A first image compression module, used for acquiring a detail image of the target denoised image, and performing dynamic range compression on the target denoised image to obtain a first compressed image; a second image compression module, configured to perform dynamic range compression on the first compressed image to obtain a second compressed image, and map the first compressed image using a gamma correction algorithm based on a lookup table method to obtain a processed image; wherein both the second compressed image and the processed image are images of a second preset bit; a first image superposition module, configured to superimpose the detail image onto the second compressed image based on different preset coefficients to obtain a first style target image and a second style target image, respectively, and to superimpose the detail image onto the processed image to obtain a third style target image; a second image superposition module, configured to perform Sobel filtering on the second compressed image in different directions to obtain filtered images in different directions, and superimpose the filtered images to obtain a fourth style target image; The target image display module is used to display the first style target image, the second style target image, the third style target image and the fourth style target image in a display interface.
9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor is used to execute the computer program to implement the steps of the multi-channel ISP processing method of infrared images as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: Used to store computer programs; wherein, when the computer program is executed by a processor, the steps of the multi-channel ISP processing method for infrared images as described in any one of claims 1 to 7 are implemented.
