Image preprocessing method and device, wearable equipment and medium

Abstract

The embodiment of the invention discloses an image preprocessing method and device, wearable equipment and a medium, and relates to the technical field of augmented reality. The method is suitable for image preprocessing in a weak light environment. The method comprises the following steps: denoising initial image information to obtain denoised image information; edge detection processing is carried out on the de-noised image information to obtain edge image information so as to extract an edge part in the image, and finally superposition processing and image enhancement processing operation are carried out on the edge image information and the de-noised image information to obtain preprocessed image information with a high signal-to-noise ratio and a clear image contour. Compared with an existing image preprocessing mode in a strong light environment or a normal light environment, image information with a high signal-to-noise ratio can be obtained through image edge extraction and superposition processing, part of lost information of the image is restored, then image information obtained through image enhancement processing is obtained, details of image features are enhanced, and the accuracy of image processing is improved. The pose positioning accuracy is improved, and the user experience is improved.

Description

Technical Field

[0001] The present application relates to the field of extended reality technology, and in particular to an image preprocessing method, apparatus, wearable device and medium. Background Art

[0002] In the field of extended reality technology, the product's posture positioning mainly relies on the cooperation between the Simultaneous Localization And Mapping (SLAM) algorithm and the Inertial Measurement Unit (IMU). Under normal lighting conditions, the collected image quality is high, and the SLAM positioning technology can maintain a high level of accuracy, giving users a good experience.

[0003] In a weak light environment, since the light in the surrounding environment is weak during the image acquisition process, the image will have a lot of noise, uneven image illumination, and even partial loss of image information. In the corresponding image preprocessing process in a normal light environment or a strong light environment, the image will be weakened for the more exposed image. If this preprocessing method is used, the light in a weak light environment is much weaker than that in a normal light environment and a strong light environment, resulting in the image being further weakened after image preprocessing, which seriously affects the accuracy of the subsequent SLAM algorithm and makes the position positioning inaccurate.

[0004] Therefore, how to improve the accuracy of posture positioning after image preprocessing is a problem that technicians in this field need to solve. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide an image preprocessing method, apparatus, wearable device and medium. Compared with the existing image preprocessing method using a strong lighting environment or a normal lighting environment, the image edge is extracted, and image information with a high signal-to-noise ratio can be obtained through superposition processing, thereby realizing the restoration of part of the lost information of the image. The image information obtained through image enhancement processing strengthens the details of the image features, improves the accuracy of posture positioning, and enhances the user experience.

[0006] In order to solve the above technical problems, the present application provides an image preprocessing method, including:

[0007] Acquire initial image information, and perform denoising processing on the initial image information to obtain denoised image information;

[0008] Performing edge detection processing on the denoised image information to obtain edge image information;

[0009] Superposition processing and image enhancement processing are performed according to the denoised image information and the edge image information to obtain pre-processed image information.

[0010] Preferably, performing superposition processing and image enhancement processing on the denoised image information and the edge image information to obtain pre-processed image information includes:

[0011] Superimposing the denoised image information and the edge image information to obtain initial superimposed image information;

[0012] Performing expansion and corrosion processing on the initial superimposed image information and the denoised image information to obtain contour image information;

[0013] The contour image information is superimposed with the initial superimposed image information to obtain the preprocessed image information.

[0014] Preferably, the step of performing expansion and corrosion processing on the initial superimposed image information and the denoised image information to obtain contour image information includes:

[0015] Performing expansion processing on the initial superimposed image information according to the denoised image information to obtain expanded image information;

[0016] Performing corrosion processing on the initial superimposed image information according to the denoised image information to obtain corrosion image information;

[0017] The dilated image information and the eroded image information are subjected to difference processing to obtain the contour image information.

[0018] Preferably, the denoising the initial image information to obtain denoised image information comprises:

[0019] The denoised image information is obtained by performing denoising processing on the initial image information according to frequency domain analysis and time domain analysis, or the time domain analysis.

[0020] Preferably, when the initial image information is subjected to denoising according to the frequency domain analysis and the time domain analysis, the process of determining the denoised image information includes:

[0021] Acquire a time domain signal of the initial image information;

[0022] Analyzing and processing the time domain signal according to the frequency domain analysis to obtain the frequency domain signal of the initial image information, wherein the frequency domain analysis at least includes one or more of spectrum analysis, energy spectrum analysis, power spectrum analysis and wavelet analysis;

[0023] Performing filtering processing on the frequency domain signal using a low-pass filter to obtain a valid signal of the frequency domain signal;

[0024] Analyzing and processing the effective information of the frequency domain signal according to the time domain analysis to obtain the denoised image information;

[0025] If the frequency domain analysis is Fourier transform analysis, the time domain analysis is the inverse transform of Fourier transform.

[0026] Preferably, performing edge detection processing on the denoised image information to obtain edge image information includes:

[0027] Traversing the pixel point information of the denoised image information to determine the gradient value of each pixel point information;

[0028] Determine the edge area and the smooth area of ​​the denoised image information according to the relationship between the gradient value of each pixel point information and the gradient threshold;

[0029] The edge image information is obtained by performing edge detection processing on the edge area and the smooth area according to a grayscale edge detection method.

[0030] Preferably, performing edge detection processing on the edge area and the smooth area according to a grayscale edge detection method to obtain the edge image information includes:

[0031] Obtaining grayscale values ​​of pixels corresponding to the edge area and the smooth area respectively;

[0032] Determining the grayscale value to be updated of the edge area according to the relationship between the gradient value of the edge area and the gain coefficient;

[0033] The grayscale value to be updated replaces the grayscale value corresponding to the pixel point in the edge area;

[0034] Update the grayscale value of the pixel points in the smooth area to 0;

[0035] The updated image information of the edge area and the smooth area is used as the edge image information.

[0036] In order to solve the above technical problems, the embodiment of the present application further provides an image preprocessing device, comprising:

[0037] An acquisition module, used for acquiring initial image information, and performing denoising processing on the initial image information to obtain denoised image information;

[0038] A first processing module, configured to perform edge detection processing on the denoised image information to obtain edge image information;

[0039] The second processing module is used to perform superposition processing and image enhancement processing according to the denoised image information and the edge image information to obtain pre-processed image information.

[0040] In order to solve the above technical problems, the embodiment of the present application further provides a wearable device, including:

[0041] Memory for storing computer programs;

[0042] A processor is used to execute the computer program to implement the steps of the image preprocessing method as described above.

[0043] In order to solve the above technical problems, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the image preprocessing method as described above are implemented.

[0044] It can be seen from the above technical scheme that an image preprocessing method provided by the present application is suitable for image preprocessing in a weak light environment. The initial image information is collected, and the initial image information is first denoised to obtain denoised image information, wherein the denoised image information is effective information, and edge details are retained while denoising, and there are no blurred details; the denoised image information is then edge detected to obtain edge image information to extract the edge part of the image; finally, the edge image information and the denoised image information are superimposed and image enhanced to obtain preprocessed image information with a high signal-to-noise ratio and clear image contours. Compared with the existing image preprocessing method using a strong light environment or a normal light environment, the image edge is extracted, and image information with a high signal-to-noise ratio can be obtained by superimposing, thereby restoring part of the lost information of the image; the image information obtained by image enhancement strengthens the details of the image features, improves the accuracy of posture positioning, and enhances the user experience.

[0045] In addition, the present application also provides an image preprocessing device, a wearable device and a medium, which have the same beneficial effects as the above-mentioned image preprocessing method. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 A flowchart of an image preprocessing method provided in an embodiment of the present application;

[0048] Figure 2 A schematic diagram of a denoising process provided in an embodiment of the present application;

[0049] Figure 3A schematic diagram of another image preprocessing provided in an embodiment of the present application;

[0050] Figure 4 A schematic diagram of the structure of an image preprocessing device provided in an embodiment of the present application;

[0051] Figure 5 A structural diagram of a wearable device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] 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 application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0053] The terms "including" and "having" in the specification and claims of this application and the above drawings, as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may include steps or units that are not listed.

[0054] The core of this application is to provide an image preprocessing method, apparatus, wearable device and medium. Compared with the existing image preprocessing method using strong lighting environment or normal lighting environment, the image edge is extracted, and image information with high signal-to-noise ratio can be obtained through superposition processing, thereby realizing the restoration of part of the lost information of the image. The image information obtained through image enhancement processing strengthens the details of the image features, improves the accuracy of posture positioning, and enhances the user experience.

[0055] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0056] It should be noted that the extended reality technology field involved in this application, Extended Reality (XR), refers to the combination of reality and virtuality through computers to create a virtual environment for human-computer interaction. This is also a general term for various technologies such as augmented reality (AR), virtual reality (VR), and mixed reality (MR). By integrating the visual interaction technologies of the three, it brings the experiencer an "immersive feeling" of seamless transition between the virtual world and the real world.

[0057] Next, an image preprocessing method provided by an embodiment of the present application is introduced in detail. Figure 1 A flowchart of an image preprocessing method provided in an embodiment of the present application, the method comprising:

[0058] S11: Acquire initial image information, and perform denoising on the initial image information to obtain denoised image information;

[0059] S12: performing edge detection processing on the denoised image information to obtain edge image information;

[0060] S13: performing superposition processing and image enhancement processing according to the denoised image information and the edge image information to obtain pre-processed image information.

[0061] It can be understood that the initial image information obtained can be image information under the lighting environment collected by the image sensor. The lighting environment can be strong light, normal light, weak light environment, etc. The image preprocessing method of this embodiment is suitable for image processing in weak light environment.

[0062] The collected initial image information is subjected to denoising to obtain denoised image information. The denoising process is to remove the interference signal. Image denoising refers to the process of reducing noise in digital images. In reality, digital images are often affected by imaging equipment and external environmental noise during digitization and transmission. The image obtained under such conditions is called a noisy image or a noisy image. Noise is an important factor that interferes with images. In practical applications, an image may contain various noises, which may be generated during the transmission process or during the quantization process. Denoising in the time domain or in the frequency domain can be used. It should be noted that the signal presentation of the collected initial image information is a time domain signal. If denoising in the frequency domain is performed, the time domain signal needs to be converted into a frequency domain signal. After the frequency domain signal processing is completed, in order to facilitate subsequent image processing in other ways, the frequency domain signal needs to be converted back into a time domain signal. Whether the denoising process is frequency domain denoising or time domain denoising, the specific denoising algorithm used is not limited and can be set according to actual conditions.

[0063] After obtaining the denoised image information, the denoised image information is subjected to edge detection processing to extract edge information and obtain edge image information. Edge detection is one of the important technologies in image processing and computer vision. Its purpose is to detect and identify a set of pixels with drastic brightness changes in an image. Correct detection of image edges plays an important role in analyzing the content in the image, segmenting and locating objects in the image, etc. Edge detection greatly reduces the amount of data in the source image, eliminates information irrelevant to the target, and retains important structural properties of the image.

[0064] The edge of an image refers to the area where the grayscale value of the pixel in the image suddenly changes. If each row and column of pixels in the image is described as a function of grayscale value, then the edge of the image corresponds to the area where the function value suddenly increases in the grayscale value function. The trend of the function value can be described by the derivative of the function. When the function value suddenly increases, the derivative will inevitably increase, while the derivative value is relatively small in the area where the function value changes more slowly. Therefore, the area with sudden changes in the function can be found by looking for areas with larger derivative values, and then the edge position in the image can be determined.

[0065] In addition, there is no limitation on the software environment used for edge detection, which can be MATLAB, Python, etc., and can be limited according to actual conditions.

[0066] The denoised image information and the edge image information are superimposed and enhanced to obtain the final preprocessed image information. The superimposition of the two image information can obtain an image with a high signal-to-noise ratio and enhanced edges and details. The higher the signal-to-noise ratio, the less noise is generated and the better the audio product. The superposition processing in this embodiment can be a single superposition or multiple superpositions, preferably multiple superpositions, where the superposition is to extract a contour image once, and the superposition is multiple times. After extracting multiple contour images, all are superimposed on the original image, so that the final superimposed image information is fuller. The details may be less visible in a single superposition, and multiple superpositions can make the effect of showing the details more obvious.

[0067] It is understandable that the image enhancement processing method is not limited to a specific image enhancement algorithm, as long as the contour image is obtained through the image enhancement method, such as expansion and corrosion, image point operation and image grayscale processing, image grayscale linear transformation, image grayscale nonlinear transformation and thresholding processing.

[0068] In addition, there is no limitation on the order of overlay processing and image enhancement processing. Overlay processing can be performed first, and then image enhancement processing, or image enhancement processing can be performed first, and then overlay processing can be performed. Overlay processing can also be interspersed with image enhancement processing, and overlay processing can be performed last, etc. There is no limitation on this, as long as the pre-processed image information has a high signal-to-noise ratio, clear contours, and can restore some lost information, and the posture positioning obtained by using this image for subsequent SLAM operations is relatively accurate.

[0069] An image preprocessing method provided in an embodiment of the present application is suitable for image preprocessing in a weak light environment. Initial image information is collected, and the initial image information is first denoised to obtain denoised image information, wherein the denoised image information is effective information, and edge details are retained while denoising, and no blurred details are contained; the denoised image information is then edge detected to obtain edge image information to extract edge portions of the image; finally, the edge image information and the denoised image information are superimposed and image enhanced to obtain preprocessed image information with a high signal-to-noise ratio and clear image contours. Compared with the existing image preprocessing method in a strong light environment or a normal light environment, the image edge is extracted, and image information with a high signal-to-noise ratio can be obtained by superimposing, thereby restoring part of the lost information of the image; the image information obtained by image enhancement strengthens the details of the image features, improves the accuracy of posture positioning, and enhances the user experience.

[0070] On the basis of the above embodiment, the step S13 performs superposition processing and image enhancement processing according to the denoised image information and the edge image information to obtain the pre-processed image information, including:

[0071] Superimposing the denoised image information and the edge image information to obtain initial superimposed image information;

[0072] Perform expansion and corrosion processing on the initial superimposed image information and the denoised image information to obtain the contour image information;

[0073] The contour image information is superimposed with the initial superimposed image information to obtain preprocessed image information.

[0074] Specifically, the denoised image information is the image g2(x, y) obtained by the denoising process in step S11, and the edge image information is the image E(x, y) obtained by the edge detection process in step S12. The two image information are superimposed to obtain the initial superimposed image information g3(x, y), and the specific formula is as follows:

[0075] g3(x,y)=g2(x,y)+E(x,y);

[0076] Then, the initial superimposed image information g3(x,y) and the denoised image information g2(x,y) are dilated and eroded to obtain the contour image information g6(x,y). Image erosion (Erosion) and dilation (Dilation) are two basic morphological operations, which are mainly used to find the minimum and maximum areas in the image. Image erosion is similar to "domain erosion". It reduces and refines the highlight area or white part of the image, and its operation result is smaller than the highlight area of ​​the original image. Image dilation is the inverse operation of the erosion operation, similar to "domain expansion". It expands the highlight area or white part of the image, and its operation result is larger than the highlight area of ​​the original image.

[0077] As an embodiment, the initial superimposed image information and the denoised image information are subjected to dilation and corrosion processing to obtain contour image information, including:

[0078] Performing expansion processing on the initial superimposed image information according to the denoised image information to obtain expanded image information;

[0079] Corrosion processing is performed on the initial superimposed image information according to the denoised image information to obtain corroded image information;

[0080] The dilated image information and the eroded image information are subtracted to obtain the contour image information.

[0081] Specifically, the denoised image information performs dilation processing and erosion processing on the initial superimposed image information to obtain the corresponding dilated image information g4(x, y) and eroded image information g5(x, y), and then performs difference processing on the dilated image information g4(x, y) and the eroded image information g5(x, y) to obtain the contour image information g6(x, y). The specific formula is as follows:

[0082] g6(x,y)=g4(x,y)-g5(x,y);

[0083] After obtaining the contour image information, a second superposition process is performed, that is, the contour image information g6 (x, y) is superimposed with the initial superposition image information g3 (x, y) to obtain the pre-processed image information g7 (x, y), and the formula is as follows:

[0084] g7(x,y)=g6(x,y)+g3(x,y);

[0085] The overlay processing and image enhancement processing provided in the embodiments of the present application can obtain image information with a high signal-to-noise ratio through overlay processing, thereby restoring some lost information of the image. The image information obtained through image enhancement processing strengthens the details of the image features, improves the accuracy of posture positioning, and enhances the user experience.

[0086] Based on the above embodiment, the step S11 of performing denoising processing on the initial image information to obtain denoised image information includes:

[0087] The initial image information is denoised according to frequency domain analysis and time domain analysis, or time domain analysis, to obtain denoised image information.

[0088] It can be understood that the initial image information is collected as a time domain signal. In the denoising process, the frequency domain method or the time domain method can be used. When the frequency domain method is used, the time domain signal needs to be converted into a frequency domain signal. After the frequency domain analysis is completed, the frequency domain signal is converted into a time domain signal to facilitate subsequent other image processing.

[0089] There is no limitation on time domain analysis and frequency domain analysis. Time domain filtering and frequency domain filtering have a certain weakening effect on image noise and edge detection effect. Time domain filtering and frequency domain filtering have their own characteristics. From the filtering effect of time domain and frequency domain low-pass filters on pictures, in time domain filtering, the smoothing filter algorithm is simple and the processing speed is fast, but it makes the image blurred while reducing noise, especially at the edges and details. The median filter has a better suppression of salt and pepper noise, but it is not suitable for images with more points and lines. Time domain low-pass filtering has a poor filtering effect on salt and pepper noise, and the image is relatively blurred. In frequency domain filtering, denoising will cause the loss of image edge information and blur the image edge, and there is a ringing effect, and the calculation amount is large and the calculation time is long. From the filtering effect of time domain and frequency domain high-pass filters on pictures, the algorithm of time domain filtering is relatively simple, the processing speed is fast, the effect is obvious in sharpening, and the lines are prominent; in frequency domain filtering, the algorithm is complex, the calculation is slow, and there is a slight ringing effect, and the image effect is relatively flat.

[0090] Frequency domain analysis does not contain blurred details, and retains image edges and details while removing noise. As an embodiment, when the initial image information is denoised according to frequency domain analysis and time domain analysis, the process of determining the denoised image information includes:

[0091] Obtaining a time domain signal of initial image information;

[0092] Analyzing and processing the time domain signal according to frequency domain analysis to obtain a frequency domain signal of the initial image information, wherein the frequency domain analysis includes at least one or more of spectrum analysis, energy spectrum analysis, power spectrum analysis and wavelet analysis;

[0093] The frequency domain signal is filtered using a low-pass filter to obtain a valid signal of the frequency domain signal;

[0094] The effective information of the frequency domain signal is analyzed and processed according to the time domain analysis to obtain the denoised image information;

[0095] Among them, if the frequency domain analysis is Fourier transform analysis, the time domain analysis is the inverse transform of Fourier transform.

[0096] This embodiment adopts frequency domain analysis, and common methods include spectrum, energy spectrum, power spectrum, inverse spectrum, wavelet analysis, etc., which are not limited here, and can be one or a combination of multiple methods. According to the frequency domain analysis, the time domain signal of the initial image information is analyzed and processed to obtain a frequency domain signal, and then the frequency domain signal is filtered through a low-pass filter to remove various types of noise in the image. The noise information of the high-frequency signal is filtered out, and the image information of the low-frequency signal is retained as valid information. It should be noted that the valid information at this time is still presented in the form of a frequency domain signal. For the subsequent image information processing, the frequency domain signal needs to be converted into a time domain signal at this time.

[0097] If the current frequency domain analysis uses Fourier transform analysis of spectrum analysis, the inverse transform of Fourier transform can be used when converting to time domain signal. Specifically, the collected digital image information is converted from time domain signal g7(t) to Fourier signal G(w), and the formula is as follows:

[0098]

[0099] Where t is time.

[0100] At this time, the Fourier signal includes a high-frequency signal G1(w) and a low-frequency signal G2(w). When designing a low-pass filter, there is no limitation on the type of low-pass filter and the filter parameters, and the filter can be designed according to the actual situation to remove various types of noise in the image. After the Fourier signal G(w) enters the low-pass filter, the high-frequency signal G1(w) is filtered out and the low-frequency signal G2(w) is retained.

[0101] It should be noted that the high-frequency signal and the low-frequency signal are determined relative to the frequency threshold, and the frequency threshold can be determined based on an empirical value, or the frequency signals can be sorted from large to small, with the first N signals as high-frequency signals, and correspondingly, the N+1th frequency signal as the frequency threshold, and the following N+1th frequency signals until the last frequency as low-frequency signals. At this time, the Fourier signal G(w) only contains the valid signal G2(w). Figure 2 A schematic diagram of a denoising process provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the time domain signal g(t) is converted into a Fourier signal G(w). After the Fourier signal G(w) enters the low-pass filter, the high-frequency signal G1(w) is filtered out and the low-frequency signal G2(w) is retained. The Fourier signal G2(w) is then inversely Fourier transformed to obtain the time domain signal g2(t) as the denoised image information, so that most of the noise in the image can be removed and the effective information can be retained.

[0102] The denoising process provided in this embodiment uses frequency domain analysis to remove most of the noise in the image and retain the effective information.

[0103] Based on the above embodiment, the step S12 of performing edge detection processing on the denoised image information to obtain edge image information includes:

[0104] Traversing the pixel information of the denoised image information to determine the gradient value of each pixel information;

[0105] Determine the edge area and the smooth area of ​​the denoised image information according to the relationship between the gradient value of each pixel point information and the first gradient threshold;

[0106] The edge detection process is performed on the edge area and the smooth area according to the grayscale edge detection method to obtain the edge image information.

[0107] Specifically, the image is edge detected through the gradient value of the image to sharpen the image information and enhance the edges and details of the image. The mathematical gradient formula is as follows:

[0108]

[0109] Since the signal of image processing is a discrete signal, the signal of denoised image information is a continuous signal. The above formula needs to be discretized. The corresponding discrete formula is as follows:

[0110] grad[f(x,y)]={[f(x,y)-f(x+1,y)] 2 +[f(x,y)-f(x,y+1)] 2} 1 / 2 ;

[0111] According to the above formula, the pixel points of the denoised image information are traversed to calculate the gradient value of each pixel point. The edge area and smooth area of ​​the denoised image information are determined according to the relationship between each gradient value and the gradient threshold. The image to which the pixel points corresponding to the gradient value greater than the gradient threshold belong is regarded as the edge area, and the image to which the pixel points corresponding to the gradient value less than or equal to the gradient threshold belong is regarded as the smooth area.

[0112] It should be noted that the gradient threshold can be determined based on an empirical value or based on the image information each time. For example, the gradient values ​​are sorted in descending order, and the first 50 gradient values ​​are used as the edge area, and the gradient threshold is the 51st gradient value. The method for selecting the gradient threshold can be determined according to mathematical rules and is not limited here.

[0113] Then, the edge detection is performed on the edge area and the smooth area according to the grayscale edge detection method to obtain edge image processing. The specific method used in the edge detection processing is not limited, and it mainly realizes strengthening the edge area and removing the smooth area.

[0114] As an embodiment, edge detection processing is performed on the edge area and the smooth area according to the grayscale edge detection method to obtain edge image information, including:

[0115] Get the grayscale values ​​of the pixels corresponding to the edge area and the smooth area respectively;

[0116] Determine the grayscale value to be updated in the edge area according to the relationship between the gradient value and the gain coefficient of the edge area;

[0117] The grayscale value to be updated replaces the grayscale value corresponding to the pixel point in the edge area;

[0118] Update the grayscale value of the pixels in the smooth area to 0;

[0119] The image information of the updated edge area and the smooth area is used as edge image information.

[0120] Specifically, the grayscale value to be updated is determined according to the gradient value and gain coefficient of the edge area, and the grayscale value corresponding to the pixel point in the edge area is replaced by the grayscale value to be updated to strengthen the edge area. At the same time, the grayscale value of the pixel point in the smooth area is updated to 0, and the formula is as follows:

[0121]

[0122] Among them, k is the gain coefficient, T is the gradient threshold, H(x, y) is the updated image grayscale value, and multiple H(x, y) values ​​constitute the grayscale value of the pixel point of the edge image information E(x, y), and the edge image information E(x, y) only contains the edge part of the image.

[0123] The edge image information is then used for processing in step 13 to obtain pre-processed image information.

[0124] Figure 3 A schematic diagram of another image preprocessing provided in an embodiment of the present application is shown in FIG. Figure 3 As shown in the figure, the initial image information is collected, and G(w) is obtained through Fourier transform, and then G2(w) is obtained through low-pass filter processing. G2(w) is subjected to inverse Fourier transform to obtain g2(t). At this time, denoising is completed, and then grayscale edge detection is performed to obtain g3(t). Finally, the pre-processed image information g7(t) is obtained through expansion and corrosion image enhancement.

[0125] The edge detection processing process provided by the embodiment of the present application strengthens the edge area of ​​the image and weakens and removes the smooth area to facilitate subsequent overlay processing and image enhancement processing.

[0126] Figure 4 A schematic diagram of the structure of an image preprocessing device provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the device comprises:

[0127] The acquisition module 11 is used to acquire the initial image information and perform denoising on the initial image information to obtain denoised image information;

[0128] The first processing module 12 is used to perform edge detection processing on the denoised image information to obtain edge image information;

[0129] The second processing module 13 is used to perform superposition processing and image enhancement processing according to the denoised image information and the edge image information to obtain pre-processed image information.

[0130] Figure 4 The description of the features in the corresponding embodiments can be found in Figure 1 The relevant descriptions of the corresponding embodiments will not be repeated here one by one.

[0131] For an introduction to an image preprocessing device provided by the present application, please refer to the above method embodiment, and the present application will not go into details here. It has the same beneficial effects as the above image preprocessing method.

[0132] Figure 5 A structural diagram of a wearable device provided in an embodiment of the present application, such as Figure 5 As shown, the electronic device includes: a memory 20 for storing a computer program;

[0133] The processor 21 is used to implement the steps of the image preprocessing method in the above embodiment when executing the computer program.

[0134] The wearable device provided in this embodiment is applicable to the fields of extended reality technologies such as virtual reality technology, augmented reality technology, mixed reality, etc., and is usually a wearable device, and its main presentation form is a terminal product, etc.

[0135] 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.

[0136] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 20 is at least used to store the following computer program 201, wherein, after the computer program is loaded and executed by the processor 21, it can implement the relevant steps of MP23015755X disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 202 may include Windows, Unix, Linux, etc. Data 203 may include, but is not limited to, image preprocessing methods, etc.

[0137] In some embodiments, the wearable device may further include a display screen 22 , an input / output interface 23 , a communication interface 24 , a power source 25 , and a communication bus 26 .

[0138] Those skilled in the art will understand that Figure 5 The structure shown in the figure does not constitute a limitation on the image preprocessing method, and may include more or fewer components than those shown in the figure.

[0139] For an introduction to a wearable device provided in the present application, please refer to the above method embodiment, which will not be repeated in the present application. It has the same beneficial effects as the above image preprocessing method.

[0140] It is understandable that if the image preprocessing method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the current technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium to execute all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, disk or optical disk and other media that can store program codes.

[0141] For an introduction to a computer-readable storage medium provided in the present application, please refer to the above method embodiment, which will not be repeated in the present application. It has the same beneficial effects as the above image preprocessing method.

[0142] The above is a detailed introduction to the image preprocessing method, device, wearable device and medium provided in the embodiments of the present application. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and 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 description.

[0143] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in 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 the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians 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.

[0144] The above is a detailed introduction to the image preprocessing method, apparatus, wearable device and medium provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. An image preprocessing method, characterized in that: include: Acquire initial image information, and perform denoising processing on the initial image information to obtain denoised image information; Performing edge detection processing on the denoised image information to obtain edge image information; Superposition processing and image enhancement processing are performed according to the denoised image information and the edge image information to obtain pre-processed image information.

2. The image preprocessing method according to claim 1, characterized in that: The performing superposition processing and image enhancement processing according to the denoised image information and the edge image information to obtain pre-processed image information includes: Superimposing the denoised image information and the edge image information to obtain initial superimposed image information; Performing expansion and corrosion processing on the initial superimposed image information and the denoised image information to obtain contour image information; The contour image information is superimposed with the initial superimposed image information to obtain the pre-processed image information.

3. The image preprocessing method according to claim 2, characterized in that: The step of performing expansion and corrosion processing on the initial superimposed image information and the denoised image information to obtain contour image information includes: Performing expansion processing on the initial superimposed image information according to the denoised image information to obtain expanded image information; Performing corrosion processing on the initial superimposed image information according to the denoised image information to obtain corrosion image information; The dilated image information and the eroded image information are subjected to difference processing to obtain the contour image information.

4. The image preprocessing method according to claim 1, characterized in that: The denoising process of the initial image information to obtain denoised image information includes: The denoised image information is obtained by performing denoising processing on the initial image information according to the frequency domain analysis and the time domain analysis, or the time domain analysis.

5. The image preprocessing method according to claim 4, characterized in that: When the initial image information is subjected to denoising according to the frequency domain analysis and the time domain analysis, the process of determining the denoised image information includes: Acquire a time domain signal of the initial image information; Analyzing and processing the time domain signal according to the frequency domain analysis to obtain the frequency domain signal of the initial image information, wherein the frequency domain analysis at least includes one or more of spectrum analysis, energy spectrum analysis, power spectrum analysis and wavelet analysis; Performing filtering processing on the frequency domain signal using a low-pass filter to obtain a valid signal of the frequency domain signal; Analyzing and processing the effective information of the frequency domain signal according to the time domain analysis to obtain the denoised image information; If the frequency domain analysis is Fourier transform analysis, the time domain analysis is the inverse transform of Fourier transform.

6. The image preprocessing method according to claim 5, characterized in that: The step of performing edge detection processing on the denoised image information to obtain edge image information includes: Traversing the pixel point information of the denoised image information to determine the gradient value of each pixel point information; Determine the edge area and the smooth area of ​​the denoised image information according to the relationship between the gradient value of each pixel point information and the gradient threshold; The edge image information is obtained by performing edge detection processing on the edge area and the smooth area according to a grayscale edge detection method.

7. The image preprocessing method according to claim 6, characterized in that: The step of performing edge detection processing on the edge area and the smooth area according to the grayscale edge detection method to obtain the edge image information includes: Obtaining grayscale values ​​of pixels corresponding to the edge area and the smooth area respectively; Determining the grayscale value to be updated of the edge area according to the relationship between the gradient value of the edge area and the gain coefficient; The grayscale value to be updated replaces the grayscale value corresponding to the pixel point in the edge area; Update the grayscale value of the pixel points in the smooth area to 0; The updated image information of the edge area and the smooth area is used as the edge image information.

8. An image preprocessing device, characterized in that: include: An acquisition module, used for acquiring initial image information, and performing denoising processing on the initial image information to obtain denoised image information; A first processing module, configured to perform edge detection processing on the denoised image information to obtain edge image information; The second processing module is used to perform superposition processing and image enhancement processing according to the denoised image information and the edge image information to obtain pre-processed image information.

9. A wearable device, characterized in that: include: Memory for storing computer programs; A processor, configured to execute the computer program to implement the steps of the image preprocessing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the image preprocessing method according to any one of claims 1 to 7 are implemented.

Images