Infrared digital holographic reconstructed image gray level equalization image enhancement method

Abstract

The invention discloses an infrared digital holographic reconstructed image grayscale equalization image enhancement method, which relates to the technical field of image processing, and comprises the steps of collecting image information, analyzing the image information, segmenting and mapping the image, and displaying the image. Obtaining reverse statistical histogram distribution of the image information so as to obtain a first intermediate image; according to a maximum between-class variance method, determining an optimal threshold value for threshold segmentation of the first intermediate image, and according to the optimal threshold value, performing threshold segmentation on the first intermediate image to obtain a second intermediate image; performing multi-segment mapping on the second intermediate image, and enhancing the image of the infrared digital holographic reconstructed image so as to obtain an enhanced final image; the distribution effect of the image in different gray level layers is improved, and the contrast of multiple areas of the image is improved to different degrees.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to an infrared digital holographic reconstruction image grayscale equalization image enhancement method. Background Art

[0002] With the development of science and technology, image enhancement technology is becoming more and more perfect. In the competition of the same type of image enhancement technology, the key and core of image enhancement is to do a good job of image layering, region division and grayscale threshold analysis, which determines the effect of image distribution in different grayscale layers. Therefore, it is very important to analyze the grayscale threshold and segmentation mapping of the image.

[0003] At present, the feasibility of image enhancement is mainly analyzed by analyzing the pixels of the image. Obviously, this analysis method has the following problems: 1. The feasibility of image enhancement is mainly analyzed by analyzing the pixels of the image, and the second intermediate image of the image is not multi-segment mapped. The optimal threshold for threshold segmentation of the first intermediate image is confirmed by the maximum inter-class variance method, and the first intermediate image is threshold segmented according to the optimal threshold to obtain the second intermediate image. As a result, the actual situation of the current image cannot be understood more accurately, and the comprehensiveness and authenticity of the image enhancement process cannot be guaranteed, and the reference and accuracy of the image enhancement results cannot be guaranteed, and the reference required for updating and upgrading of subsequent methods cannot be provided.

[0004] 2. Currently, there is no specific calculation formula to enhance the infrared digital holographic reconstructed image, and the final image is not displayed. It is impossible to truly show the enhancement of the final image, and it is impossible to provide the background pixel area of ​​the compressed high-bit image and the expansion of the main detail pixel area, and thus it is impossible to guarantee the image enhancement, thereby failing to improve the efficiency of the infrared digital holographic reconstructed image enhancement, and to a certain extent reducing the image enhancement effect of the infrared digital holographic reconstructed image. Summary of the invention

[0005] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide an infrared digital holographic reconstructed image grayscale equalization image enhancement method.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions: In the first aspect, the present invention provides an infrared digital holographic reconstruction image grayscale equalization image enhancement method, the method comprising the following steps: Step 1, collecting image information: encoding the wavefront information of the object through an optical element, converting it into a planar interference pattern, irradiating the encoded wavefront information to a photodetector to convert it into an electrical signal, and transmitting it to the interference record; using an analog-to-digital converter to convert the electrical signal of the interference record into a digital signal, and then using a back propagation algorithm to reconstruct the digital signal to obtain the infrared radiation distribution of the original object; holographically encoding the reconstructed infrared radiation distribution to obtain a hologram, using a variational autoencoder to reconstruct the infrared radiation distribution, and collecting the corresponding image information after the image reconstruction, which is recorded as the image information of the infrared digital holographic reconstructed image.

[0007] Step 2: Analyze image information: extract image information of the infrared digital holographic reconstructed image, and then use statistical calculation to obtain the inverse statistical histogram distribution of the image information, and obtain the first intermediate image.

[0008] Step 3, image segmentation and mapping: extract the first intermediate image, confirm the optimal threshold for threshold segmentation of the first intermediate image by the maximum inter-class variance method, then perform threshold segmentation on the first intermediate image according to the optimal threshold to obtain the second intermediate image, obtain the bit width of the second intermediate image, perform multi-segment mapping on the second intermediate image, complete the image enhancement of the infrared digital holographic reconstructed image, and obtain the final image.

[0009] Step 4: Image display: When the image enhancement of the infrared digital holographic reconstructed image is completed, the final image is displayed.

[0010] Preferably, the image information of the infrared digital holographic reconstructed image includes the total number of pixels and grayscale levels of the image.

[0011] Preferably, the inverse statistical histogram distribution of the image information is obtained by using statistical calculation analysis, and the first intermediate image is obtained. The specific calculation process is as follows: the total number of pixels and the grayscale level of the image are recorded as n and L respectively, and substituted into the statistical function of the inverse statistical histogram The inverse statistical histogram distribution of the image information is obtained, where k is the function corresponding to the grayscale level in the statistical function representing the inverse statistical histogram. Expressed as a decreasing function of the inverse statistical histogram, Represented as a decreasing function of the inverse statistical histogram The next decreasing function of the image information is transmitted to the image analysis module, thereby obtaining the first intermediate image.

[0012] Preferably, the first intermediate image is threshold segmented according to the optimal threshold to obtain the second intermediate image, and the bit width of the second intermediate image is obtained. The specific analysis process is as follows: a cluster analysis method is used to obtain each characteristic grayscale value in the first intermediate image and the threshold value corresponding to each characteristic grayscale value, so as to confirm the optimal threshold for threshold segmentation of the first intermediate image by the maximum inter-class variance method; when the maximum inter-class variance function is unimodal, the threshold of the characteristic grayscale value corresponding to the peak is the optimal threshold; when the maximum inter-class variance function is bimodal or multi-modal, the threshold of the characteristic grayscale value corresponding to the valley value before the second peak is the optimal threshold; and then the first intermediate image is threshold segmented according to the optimal threshold to obtain the second intermediate image; the second intermediate image is transmitted to the image analysis module, and then the bit width of the second intermediate image is obtained.

[0013] Preferably, the second intermediate image is subjected to multi-segment mapping, and the specific mapping method is as follows: the corresponding coordinate axes of the mapped image are set to be expressed as , record the data width of the second intermediate image as m bits, and use multi-segment mapping to convert the second intermediate image with m bits of data width into an image with n bits of data width, , substitute into the corresponding conversion function Get an image with n-bit data width ,in, represents the synthetic image after segmentation according to the threshold segmentation formula, and The second intermediate image representing the m-bit data width is based on Each image divided after secondary threshold segmentation, and Indicates the corresponding threshold to be set in an image with n-bit data width.

[0014] Preferably, the segmentation is performed according to the threshold segmentation formula to obtain the final image. The specific segmentation calculation process is as follows: According to the calculation formula Get the peak and crest , A statistical function representing an inverse statistical histogram.

[0015] Preferably, the image enhancement of the infrared digital holographic reconstructed image is specifically performed as follows: According to the calculation formula The background pixel area of ​​the compressed high-bit image and the expanded main detail pixel area are obtained to complete the image enhancement of the infrared digital holographic reconstructed image and obtain the final image.

[0016] The beneficial effects of the present invention are as follows: 1. A grayscale equalization image enhancement method for an infrared digital holographic reconstructed image provided by the present invention performs statistical calculations on the image information of the collected infrared digital holographic reconstructed image, thereby obtaining the inverse statistical histogram distribution of the image information, thereby obtaining a first intermediate image; confirming the optimal threshold for threshold segmentation of the first intermediate image according to the maximum inter-class variance method, and performing threshold segmentation on the first intermediate image according to the optimal threshold to obtain a second intermediate image; performing multi-segment mapping on the second intermediate image, and then enhancing the image of the infrared digital holographic reconstructed image, thereby obtaining an enhanced final image; improving the effect of image distribution in different grayscale layers, so that the contrast of multiple regions of the image is improved to varying degrees.

[0017] 2. The present invention performs multi-segment mapping on the second intermediate image of the image, confirms the optimal threshold for threshold segmentation of the first intermediate image by the maximum inter-class variance method, and obtains the second intermediate image by threshold segmentation of the first intermediate image according to the optimal threshold, so as to more accurately understand the actual situation of the current image, while also ensuring the comprehensiveness and authenticity of the image enhancement process, thereby ensuring the reference and accuracy of the image enhancement results, and also providing the reference required for updating and upgrading of subsequent methods.

[0018] 3. The present invention enhances the image of the infrared digital holographic reconstructed image through a specific calculation formula, displays the final image, truly shows the enhancement of the final image, provides background pixel area compression of the high-bit image and expansion of the main detail pixel area, thereby ensuring image enhancement, thereby improving the efficiency of image enhancement of the infrared digital holographic reconstructed image, and improving the image enhancement effect of the infrared digital holographic reconstructed image to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] 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 work.

[0020] Figure 1 The present invention is a flowchart of the steps for implementing the method. DETAILED DESCRIPTION

[0021] 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.

[0022] See also Figure 1 As shown, the present invention provides, in a first aspect, an infrared digital holographic reconstruction image grayscale equalization image enhancement method, including the step of one, collecting image information: encoding the wavefront information of the object through an optical element, converting it into a planar interference pattern, irradiating the encoded wavefront information to a photodetector to convert it into an electrical signal, and transmitting it to the interference record; using an analog-to-digital converter to convert the electrical signal of the interference record into a digital signal, and then using a back propagation algorithm to reconstruct the digital signal to obtain the infrared radiation distribution of the original object; holographically encoding the reconstructed infrared radiation distribution to obtain a hologram, using a variational autoencoder to reconstruct the infrared radiation distribution, and collecting the image information corresponding to the image reconstruction, which is recorded as the image information of the infrared digital holographic reconstructed image.

[0023] In a specific example, the image information of the infrared digital holographic reconstructed image includes the total number of pixels and grayscale levels of the image.

[0024] It should be noted that the optical element may be a phase mask and an aspherical mirror, etc. These optical elements may be used to generate light with a certain shape and distribution so as to encode the wavefront information of the object in subsequent processing.

[0025] It should also be noted that the total number of pixels of an image is obtained by multiplying the width and height of the image; for example, if the width of an image is 1000 pixels and the height is 800 pixels, then the total number of pixels of the sub-image is 1000*800=800000 pixels; the grayscale is used to indicate the number of gray levels available for pixels in a grayscale image.

[0026] It still needs to be explained that a photodetector is a device that can detect light radiation and convert it into a processable electrical signal. It uses the characteristics of light waves radiated or reflected by the detected target, combined with changes in the conductivity characteristics of electro-optical response materials or devices, to complete the conversion and real-time mapping of optical signals to electrical signals, and then achieve accurate target detection and identification; an analog-to-digital converter is an electronic component that converts analog signals into digital signals; a variational autoencoder is mainly composed of two parts, an encoder and a decoder. The encoder compresses the input data into a latent space representation, and the decoder recovers similar input data from this representation.

[0027] Step 2: Analyze image information: extract image information of the infrared digital holographic reconstructed image, and then use statistical calculation to obtain the inverse statistical histogram distribution of the image information, and obtain the first intermediate image.

[0028] In a specific example, the inverse statistical histogram distribution of the image information is obtained by using statistical calculation analysis, and the first intermediate image is obtained. The specific calculation process is as follows: the total number of pixels and the grayscale level of the image are recorded as n and L respectively, and substituted into the statistical function of the inverse statistical histogram The inverse statistical histogram distribution of the image information is obtained, where k is the function corresponding to the grayscale level in the statistical function representing the inverse statistical histogram. Expressed as a decreasing function of the inverse statistical histogram, Represented as a decreasing function of the inverse statistical histogram The next decreasing function of the image information is transmitted to the image analysis module, thereby obtaining the first intermediate image.

[0029] It should be noted that the inverse statistical histogram distribution is a method for analyzing the distribution of image pixel intensity, which can clearly show the brightness and contrast of the image.

[0030] Step 3, image segmentation and mapping: extract the first intermediate image, confirm the optimal threshold for threshold segmentation of the first intermediate image by the maximum inter-class variance method, then perform threshold segmentation on the first intermediate image according to the optimal threshold to obtain the second intermediate image, obtain the bit width of the second intermediate image, perform multi-segment mapping on the second intermediate image, complete the image enhancement of the infrared digital holographic reconstructed image, and obtain the final image.

[0031] It should be noted that the maximum inter-class variance method is the OTSU algorithm, which is the best algorithm for threshold selection in image segmentation. It is simple to calculate and is not affected by image brightness and contrast.

[0032] In a specific embodiment, the first intermediate image is threshold segmented according to the optimal threshold to obtain the second intermediate image, and the bit width of the second intermediate image is obtained. The specific analysis process is as follows: a cluster analysis method is used to obtain each characteristic grayscale value in the first intermediate image and the threshold value corresponding to each characteristic grayscale value, so as to confirm the optimal threshold for threshold segmentation of the first intermediate image through the maximum inter-class variance method; when the maximum inter-class variance function is unimodal, the threshold of the characteristic grayscale value corresponding to the peak is the optimal threshold; when the maximum inter-class variance function is bimodal or multi-modal, the threshold of the characteristic grayscale value corresponding to the valley value before the second peak is the optimal threshold; then the first intermediate image is threshold segmented according to the optimal threshold to obtain the second intermediate image; the second intermediate image is transmitted to the image analysis module, and then the bit width of the second intermediate image is obtained.

[0033] It should be noted that the larger the inter-class variance between the background and the foreground, the greater the difference between the two parts that make up the image. When part of the foreground is mistakenly classified as the background or part of the background is mistakenly classified as the foreground, the difference between the two parts will become smaller. Therefore, the segmentation that maximizes the inter-class variance can reduce the probability of misclassification.

[0034] In a specific embodiment, the second intermediate image is subjected to multi-segment mapping, and the specific mapping method is as follows: the corresponding coordinate axis of the mapped image is set to be expressed as , record the data width of the second intermediate image as m bits, and use multi-segment mapping to convert the second intermediate image with m bits of data width into an image with n bits of data width, , substitute into the corresponding conversion function Get an image with n-bit data width ,in, represents the synthetic image after segmentation according to the threshold segmentation formula, and The second intermediate image representing the m-bit data width is based on Each image divided after secondary threshold segmentation, and Indicates the corresponding threshold to be set in an image with n-bit data width.

[0035] It should be noted that, for example, the second intermediate image with a 16-bit data width is mapped to an image with an 8-bit data width by multi-segment mapping, and the corresponding conversion function is Resulting in an image with 8-bit data width .

[0036] The present invention performs multi-segment mapping on the second intermediate image of the image, confirms the optimal threshold for threshold segmentation of the first intermediate image through the maximum inter-class variance method, and obtains the second intermediate image by threshold segmentation of the first intermediate image according to the optimal threshold, so as to more accurately understand the actual situation of the current image, while also ensuring the comprehensiveness and authenticity of the image enhancement process, thereby ensuring the reference and accuracy of the image enhancement results, and also providing the reference required for updating and upgrading of subsequent methods.

[0037] In a specific embodiment, the segmentation is performed according to the threshold segmentation formula, and the final image is obtained. The specific segmentation calculation process is as follows: According to the calculation formula Get the peak and crest , A statistical function representing an inverse statistical histogram.

[0038] In another specific embodiment, the image enhancement of the infrared digital holographic reconstructed image is specifically performed as follows: according to the calculation formula The background pixel area of ​​the compressed high-bit image and the expanded main detail pixel area are obtained to complete the image enhancement of the infrared digital holographic reconstructed image and obtain the final image.

[0039] It should be noted that the background pixel area of ​​the high-bit image is compressed, firstly because the high-bit image contains a large amount of background pixel information, and compressing these areas can significantly reduce the size of the image file and save storage space; secondly, in network transmission, compressing the image can reduce transmission time and improve transmission efficiency; secondly, in some applications, only the foreground information in the image may be needed, and compressing the background pixels can highlight the foreground and better meet application needs; finally, on some real-time processing or low-computing power devices, compressing the image can reduce processing time and computing burden.

[0040] It should also be noted that expanding the main detail pixel area can make important details in the image clearer and finer, enhance the human eye's perception and understanding of details, improve the quality of the image, and display the image's texture, color, light and shadow effects to a deeper degree.

[0041] The present invention enhances the image of the infrared digital holographic reconstructed image through a specific calculation formula, displays the final image, truly shows the enhancement of the final image, provides background pixel areas for compressing high-bit images and expanding main detail pixel areas, thereby ensuring image enhancement, thereby improving the efficiency of image enhancement of the infrared digital holographic reconstructed image, and improving the image enhancement effect of the infrared digital holographic reconstructed image to a certain extent.

[0042] Step 4: Image display: When the image enhancement of the infrared digital holographic reconstructed image is completed, the final image is displayed.

[0043] The present invention provides an infrared digital holographic reconstructed image grayscale equalization image enhancement method, which performs statistical calculation on the image information of the collected infrared digital holographic reconstructed image, and then obtains the reverse statistical histogram distribution of the image information, thereby obtaining a first intermediate image; confirms the optimal threshold for threshold segmentation of the first intermediate image according to the maximum inter-class variance method, and performs threshold segmentation on the first intermediate image according to the optimal threshold to obtain a second intermediate image; performs multi-segment mapping on the second intermediate image, and then enhances the infrared digital holographic reconstructed image, thereby obtaining an enhanced final image; improves the effect of image distribution in different grayscale layers, so that the contrast of multiple regions of the image is improved to different degrees.

[0044] The above contents are merely examples and explanations of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A method for enhancing grayscale equalization of infrared digital holographic reconstruction images, characterized in that: include: Step 1, collecting image information: encoding the wavefront information of the object through optical elements, converting it into a plane interference pattern, irradiating the encoded wavefront information to a photodetector to convert it into an electrical signal, and transmitting it to the interference record; using an analog-to-digital converter to convert the electrical signal of the interference record into a digital signal, and then using a back propagation algorithm to reconstruct the digital signal to obtain the infrared radiation distribution of the original object; holographically encoding the reconstructed infrared radiation distribution to obtain a hologram, using a variational autoencoder to reconstruct the infrared radiation distribution, and collecting the image information corresponding to the image reconstruction, which is recorded as the image information of the infrared digital holographic reconstructed image; Step 2: Analyze image information: extract image information of the infrared digital holographic reconstructed image, and then use statistical calculation to obtain the reverse statistical histogram distribution of the image information, and obtain the first intermediate image; Step 3, image segmentation and mapping: extract the first intermediate image, determine the optimal threshold for threshold segmentation of the first intermediate image by using the maximum inter-class variance method, then perform threshold segmentation on the first intermediate image according to the optimal threshold to obtain a second intermediate image, obtain the bit width of the second intermediate image, perform multi-segment mapping on the second intermediate image, complete image enhancement of the infrared digital holographic reconstructed image, and obtain the final image; Step 4: Image display: When the image enhancement of the infrared digital holographic reconstructed image is completed, the final image is displayed.

2. The method for enhancing the grayscale equalization of an infrared digital holographic reconstruction image according to claim 1, characterized in that: The image information of the infrared digital holographic reconstructed image includes the total number of pixels and grayscale levels of the image.

3. The method for enhancing the grayscale equalization of an infrared digital holographic reconstruction image according to claim 2, characterized in that: The statistical calculation analysis is used to obtain the reverse statistical histogram distribution of the image information and obtain the first intermediate image. The specific calculation process is as follows: The total number of pixels and grayscale levels of the image are recorded as n and L respectively, and substituted into the statistical function of the reverse statistical histogram The inverse statistical histogram distribution of the image information is obtained, where k is the function corresponding to the grayscale level in the statistical function representing the inverse statistical histogram. Expressed as a decreasing function of the inverse statistical histogram, Represented as a decreasing function of the inverse statistical histogram The next decreasing function of the image information is transmitted to the image analysis module, thereby obtaining the first intermediate image.

4. The method for enhancing the grayscale equalization of an infrared digital holographic reconstruction image according to claim 3, characterized in that: The first intermediate image is threshold segmented according to the optimal threshold to obtain a second intermediate image, and the bit width of the second intermediate image is obtained. The specific analysis process is as follows: A cluster analysis method is used to obtain the characteristic grayscale values ​​in the first intermediate image and the threshold values ​​corresponding to the characteristic grayscale values, so as to confirm the optimal threshold value for threshold segmentation of the first intermediate image through the maximum inter-class variance method; when the maximum inter-class variance function is unimodal, the threshold value of the characteristic grayscale value corresponding to the peak is the optimal threshold value; when the maximum inter-class variance function is bimodal or multi-modal, the threshold value of the characteristic grayscale value corresponding to the valley value before the second peak is the optimal threshold value; then the first intermediate image is threshold segmented according to the optimal threshold value to obtain a second intermediate image; the second intermediate image is transmitted to the image analysis module to obtain the bit width of the second intermediate image.

5. The method for enhancing the grayscale equalization of an infrared digital holographic reconstruction image according to claim 4, characterized in that: The multi-segment mapping is performed on the second intermediate image, and the specific mapping method is as follows: The corresponding coordinate axis of the mapped image is expressed as , record the data width of the second intermediate image as m bits, and use multi-segment mapping to convert the second intermediate image with m bits of data width into an image with n bits of data width, , substitute into the corresponding conversion function Get an image with n-bit data width ,in, represents the synthetic image after segmentation according to the threshold segmentation formula, and The second intermediate image representing the m-bit data width is based on Each image divided after secondary threshold segmentation, and Indicates the corresponding threshold to be set in an image with n-bit data width.

6. The method for enhancing the grayscale equalization of an infrared digital holographic reconstruction image according to claim 3, characterized in that: The segmentation is performed according to the threshold segmentation formula to obtain the final image. The specific segmentation calculation process is as follows: According to the calculation formula Get the peak and crest , A statistical function representing an inverse statistical histogram.

7. The method for enhancing the grayscale equalization of an infrared digital holographic reconstruction image according to claim 5, characterized in that: The specific enhancement process of the image enhancement of the infrared digital holographic reconstructed image is as follows: According to the calculation formula The background pixel area of ​​the compressed high-bit image and the expanded main detail pixel area are obtained to complete the image enhancement of the infrared digital holographic reconstructed image and obtain the final image.

Images