Random computing edge detection enhancement method, system, medium, and device

Abstract

A random computation edge detection enhancement method, system, medium and equipment based on random resonance, in the method, an image is collected and taken as an input image, the input image is converted into a gray image, and a gray value matrix is obtained by normalizing the gray value; a pulse excitation is applied, a random bit stream is generated by random flipping of a memory, and each gray value in the gray value matrix is represented by using the random bit stream; random computation is performed on the random bit stream to obtain an edge detection result; a noise signal generated during the operation of an electronic device used for collecting the image is captured, and a random bit stream with a predetermined probability is generated by sampling; and the noise random bit stream with the predetermined probability is injected into the edge detection result to achieve enhancement and optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of integrated circuits, in particular to a random computing edge detection enhancement method, system, medium and equipment based on stochastic resonance. BACKGROUND

[0002] Edge detection is a basic and key algorithm in computer vision. In an image, an edge refers to the boundary between different regions, usually containing the outline and detail information of an object. The goal of an edge detection algorithm is to find these edges in an image and extract them, providing a foundation for subsequent image analysis and processing. Edge detection has wide applications in image segmentation, object detection, and video object segmentation.

[0003] However, due to many factors, including low contrast of the image, complex background, blurred edges, etc., the edge detection result error is large, and the image edges cannot be detected and extracted. This brings great challenges to edge detection in practical applications. For example, when processing road markings and obstacles in autonomous vehicles, if the edge detection is not accurate, it may affect the judgment and safety of the vehicle. In addition, in medical diagnostic images, accurate edge detection is crucial for correctly identifying lesion areas, and false edge detection may lead to misdiagnosis.

[0004] The existing method for optimizing edge detection results mainly pre-processes the input image, enhances the image contrast, and uses image denoising, image segmentation, background modeling, etc. to process images with complex backgrounds. However, implementing the above pre-processing in a hardware-based machine learning algorithm chip requires designing complex peripheral circuits, which will bring additional circuit area and energy consumption.

[0005] Stochastic resonance is a signal enhancement phenomenon. In signal analysis, noise will reduce the signal-to-noise ratio and affect the extraction of useful information. However, when a weak signal passes through a nonlinear system, a certain matching relationship is achieved between the nonlinear system, the signal and the noise, and stochastic resonance occurs. The presence of noise increases the strength of the original weak signal, providing a new way for the detection of weak signals.

[0006] The above information disclosed in the background section is only intended to enhance the understanding of the background of the present application, and therefore can contain information that is not prior art known to those of ordinary skill in the art. SUMMARY

[0007] The present application provides a random computing edge detection enhancement method, system, medium and equipment based on stochastic resonance. The present disclosure utilizes the noise generated by the memory and is based on the principle of stochastic resonance, which can effectively overcome the problem of large error.

[0008] A random computing edge detection enhancement method based on stochastic resonance comprises:

[0009] S100: collecting an image as an input image, converting the input image into a gray image, and performing normalization processing on a gray value to obtain a gray value matrix;

[0010] S200: applying a pulse excitation, storing a random flip to generate a random bit stream, and using the random bit stream to represent each gray value in the gray value matrix;

[0011] S300: performing random computing on the random bit stream to obtain an edge detection result;

[0012] S400: capturing a noise signal generated by an electronic device during image acquisition, and sampling to generate a random bit stream with a predetermined probability;

[0013] S500: injecting the noise random bit stream with the predetermined probability into the edge detection result to achieve enhancement optimization.

[0014] In the random computing edge detection enhancement method based on stochastic resonance, in step S200, the pulse excitation is a two-stage pulse, including a probability setting pulse for causing the memory to flip with a certain probability and a reset pulse for causing the memory to return to the initial state within a period.

[0015] In the random computing edge detection enhancement method based on stochastic resonance, in step S200, the state of the memory is compared with a reference resistance, and bit 1 is generated in a high resistance state and bit 0 is generated in a low resistance state, and the probability of bit 1 in the random bit stream represents the gray value.

[0016] In the random computing edge detection enhancement method based on stochastic resonance, in step S200, the memory includes a magnetic random memory, a phase change memory, and a ferroelectric memory.

[0017] In the random computing edge detection enhancement method based on stochastic resonance, in step S400, the noise signal includes any one or any combination of the following: a random signal generated by the memory, 1 / f noise, RTN noise, and Gaussian noise.

[0018] In the random computing edge detection enhancement method based on stochastic resonance, in step S400, the random flip probability of the memory and the threshold of the electronic device noise sampling are controlled to regulate the noise signal to output a random bit stream with a predetermined probability.

[0019] In the random computing edge detection enhancement method based on stochastic resonance, the electronic device includes an autonomous vehicle for identifying road signs and obstacles or a medical diagnostic imaging device.

[0020] A random computing edge detection enhancement system comprises,

[0021] An image input unit is configured to collect an input image and convert the input image into a gray scale image, normalize pixels to obtain a gray scale value matrix;

[0022] A pulse excitation application unit is configured to apply pulse excitation to the memory;

[0023] A random bit stream generation module is configured to generate a random bit stream representing each gray scale value in the gray scale value matrix by storing the memory to randomly flip under the pulse excitation;

[0024] An edge detection random computing unit is configured to perform random computation on the random bit stream of the predetermined gray scale value to obtain an edge detection result;

[0025] A noise capturing unit is configured to capture a noise signal generated by an electronic device when the electronic device is working to collect an image;

[0026] A noise control unit is configured to control and adjust the noise signal to obtain a noise signal representing different random bit streams.

[0027] A computer storage medium comprises computer instructions, which, when executed on a computer, cause the computer to perform the method.

[0028] An electronic device comprises:

[0029] A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein,

[0030] The processor implements the method when executing the program.

[0031] Compared with the prior art, the present application has the following advantages: the present application uses the memory to generate a random bit stream for random computation, thereby obtaining an edge detection result. Meanwhile, the method further extracts a noise signal of the memory and injects the noise signal into the edge detection result to achieve enhancement and optimization of the result. By using random resonance and the random flipping characteristics of the memory, the problem of large error of the edge detection method can be effectively solved, and more accurate and reliable edge detection results can be provided. The noise generated when the memory is working is extracted and injected into the edge detection result by addition, without the need for complex peripheral circuits, thereby greatly saving hardware area and power consumption, and providing a new hardware solution for enhancement and optimization of the edge detection task. By controlling the probability of memory flipping and the threshold of sampling, the noise signal is regulated to output a random bit stream with a specific probability, different noise signals are injected into the random computing edge detection result to obtain different optimization effects, and the optimization effects are easy to regulate. BRIEF DESCRIPTION OF DRAWINGS

[0032] Various other advantages and benefits of the present application will become apparent to those of ordinary skill in the art, upon reading the following detailed description of the preferred embodiment. The accompanying drawings are included to provide a better understanding of the preferred embodiment, and are not intended to limit the scope of the application. It should be readily understood that the drawings are not to scale, and are merely intended to depict the various embodiments of the application. As such, the drawings should not be construed as limiting the scope of the application.

[0033] In the drawings:

[0034] Figure 1 Flow chart of the random resonance based random computing edge detection enhancement method of the present disclosure;

[0035] Figure 2 Graph showing the relationship between the sampling threshold and the probability of the random bit stream generated by the noise signal for another embodiment of the present disclosure;

[0036] Figure 3 Graph showing the relationship between the pulse amplitude and the probability of the random bit stream generated by the noise signal for another embodiment of the present disclosure;

[0037] Figure 4 Graph showing the optimization effect of the edge detection enhancement for another embodiment of the present disclosure;

[0038] Figure 5 Graph showing the random resonance based random computing edge detection enhancement system for another embodiment of the present disclosure.

[0039] The present application will be further explained with reference to the drawings and embodiments. DETAILED DESCRIPTION

[0040] Embodiments of the present application will be described more fully hereinafter with reference to the accompanying drawings. This application may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.

[0041] It should be noted that some terms are used in the description and claims to refer to certain components. Those skilled in the art will understand that the same component can be referred to by different terms. The description and claims do not distinguish components based on the difference in terminology, but rather on the difference in function. As mentioned throughout the description and claims, "including" or "comprising" is an open term, which should be interpreted to mean "including but not limited to". The subsequent description is a preferred embodiment of implementing the present application, which is for the purpose of illustrating the general principles of the application, and is not intended to limit the scope of the application. The scope of protection of the present application is defined by the appended claims.

[0042] In order to facilitate the understanding of the embodiments of the present application, the following will be further explained and described with specific examples combined with the accompanying drawings, and each drawing does not constitute a limitation on the embodiments of the present application.

[0043] As shown in the figure, the random computation edge detection enhancement method based on random resonance includes the following steps: Figures 1 to 5

[0044] S100: Collecting an image as an input image, converting the input image into a gray-scale image, and performing normalization processing on the gray-scale value to obtain a gray-scale value matrix;

[0045] S200: Applying a pulse excitation, storing a random flip to generate a random bit stream, and using the random bit stream to represent each gray-scale value in the gray-scale value matrix;

[0046] S300: Random computation of the random bit stream to obtain an edge detection result;

[0047] S400: Capturing the noise signal generated by the electronic device during operation for collecting the image, and sampling to generate a random bit stream with a predetermined probability;

[0048] S500: Injecting the noise random bit stream with a predetermined probability into the edge detection result to achieve enhancement optimization.

[0049] In the preferred embodiment of the random computation edge detection enhancement method based on random resonance, in step S200, the pulse excitation is a two-stage pulse, which includes a probability setting pulse for making the memory flip with a certain probability and a reset pulse for making the memory return to the initial state in one cycle.

[0050] In the preferred embodiment of the random computation edge detection enhancement method based on random resonance, in step S200, the state of the memory is compared with the reference resistance, and bit 1 is generated in the high resistance state and bit 0 is generated in the low resistance state, and the probability of bit 1 in the random bit stream represents the gray-scale value. ​

[0051] In the preferred embodiment of the random computation edge detection enhancement method based on random resonance, in step S200, the memory includes a magnetic random memory, a phase change memory, and a ferroelectric memory.

[0052] In the preferred embodiment of the random computation edge detection enhancement method based on random resonance, in step S400, the noise signal includes a random signal generated by the memory, 1 / f noise, RTN noise, and Gaussian noise.

[0053] In the preferred embodiment of the random computation edge detection enhancement method based on random resonance, in step S400, the noise signal is adjusted by controlling the random flipping probability of the memory and the threshold of the electronic device noise sampling to output a random bit stream with a predetermined probability.

[0054] In the preferred embodiment of the random computation edge detection enhancement method based on random resonance, the electronic device includes an autonomous vehicle for identifying road signs and obstacles or a medical diagnostic imaging device.

[0055] A random computation edge detection enhancement system includes,

[0056] An image input unit for collecting an input image and converting it into a grayscale image, normalizing the pixels to obtain a grayscale value matrix;

[0057] A pulse excitation application unit for applying pulse excitation to the memory;

[0058] A random bit stream generation module for the memory to generate a random bit stream representing each grayscale value in the grayscale value matrix under the pulse excitation;

[0059] An edge detection random computation unit for performing random computation on the random bit stream of the predetermined grayscale value to obtain an edge detection result;

[0060] A noise capturing unit for capturing a noise signal generated by an electronic device during image acquisition;

[0061] A noise control unit for controlling and adjusting the noise signal to obtain a noise signal representing different random bit streams.

[0062] A computer storage medium, the storage medium comprising computer instructions, when executed on a computer, causing the computer to perform the method.

[0063] An electronic device, the electronic device comprising:

[0064] The application discloses a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein,

[0065] The processor implements the method when executing the program.

[0066] In one embodiment, the noise signal is converted into a random bit stream and then injected into the edge detection result by addition.

[0067] In one embodiment, as shown in the figure, the application provides a random computing edge detection enhancement method based on random resonance, which comprises the following steps: Figure 1

[0068] S100: converting an input image into a gray-scale image, performing normalization processing on a gray-scale value to obtain a gray-scale value matrix;

[0069] The gray-scale image is composed of single pixel matrices, and each pixel matrix has an integer value between 0 and 255. The gray-scale value is normalized, that is, the gray-scale value is scaled to between 0 and 1.

[0070] S200: applying a pulse excitation, storing a random flip of a memory to generate a random bit stream, and using the random bit stream to represent each gray-scale value in the pixel matrix;

[0071] In this step, the pulse excitation applied to the memory is current excitation or voltage excitation, and the voltage excitation or current excitation is a second-order excitation in a cycle, including a probability setting pulse and a refresh pulse. The probability setting pulse can make the memory randomly flip from a low resistance state (P state) to a high resistance state (AP state) with a certain probability or randomly flip from the high resistance state (AP state) to the low resistance state (P state) with a certain probability. The refresh pulse can make the memory flip from the existing state to the low resistance state (P state) or flip from the existing state to the high resistance state (AP state). The state of the memory is compared with a reference resistance value, and a bit "1" is generated in the high resistance state and a bit "0" is generated in the low resistance state.

[0072] The normalized gray-scale value is a decimal number between 0 and 1, and the gray-scale value is represented by the probability of "1" in the random bit stream.

[0073] The memory includes a magnetic random memory, a phase change memory, a ferroelectric memory and the like.

[0074] In the preferred embodiment, the magnetic random memory has a sandwich structure, which includes a free layer, a tunneling layer and a pinning layer.

[0075] ​In the preferred embodiment, the free layer and the pinned layer are made of ferromagnetic material, including NiFe, CoFe or CoFeB; the tunneling layer is made of non-magnetic insulating material, including MgO, Al2O3, Al2MgO4, ZnO, HfO2 or TaO2.

[0076] S300: Randomly calculate the random bit stream to obtain an edge detection result.

[0077] The random calculation includes random calculation of multiplication, random calculation of addition, etc. The edge detection is performed by using a template as a kernel to perform convolution with each pixel point of the image and random calculation, and then selecting a suitable threshold to extract the edge of the image. Common templates include Roberts operator, Sobel operator and Prewitt operator. In another embodiment, the present disclosure uses Roberts operator to perform edge detection calculation.

[0078] S400: Capture the noise signal generated by the memory during operation to generate a random bit stream with a specific probability.

[0079] The noise signal includes, but is not limited to, a random signal generated by the memory, 1 / f noise, RTN noise and Gaussian noise.

[0080] By controlling the random flipping probability of the memory and the threshold value of the electronic device noise sampling, the noise signal is regulated to output a random bit stream with a specific probability.

[0081] In this embodiment, the present disclosure uses the Gaussian noise generated by the electronic device during operation as the noise signal, compares the noise signal with the threshold value H, and outputs bit "1" if it is greater than H, and outputs bit "0" if it is less than H. By regulating the threshold value, the noise signal is generated to control the bit stream with a specific probability. The relationship between the threshold value H and the probability x of the output bit stream is shown in the following formula:

[0082]

[0083] wherein, is the mean of the Gaussian noise, is the variance of the Gaussian noise. As shown in Figure 2 , the relationship between H and x is shown in the following formula:

[0084] In another embodiment, the present disclosure uses the random flipping behavior of the memory under pulse excitation as the noise signal. The amplitude of the excitation can be controlled to regulate the flipping probability, as shown in Figure 3 ​As shown, the flipping probability of the memory has a sigmoid dependence on the applied pulse excitation amplitude, and by regulating the flipping probability, the generation of the noise signal with a specific probability of the bit stream is controlled.

[0085] S500: Injecting the noise random bit stream with a specific probability into the edge detection result to achieve enhanced optimization effect.

[0086] In this embodiment, as shown, Figure 4 The image edge after edge detection of the input image is not clear and cannot be distinguished, and white noise signals with bit stream probabilities of 0.01, 0.08, 0.1 and 0.5 are applied to it in turn. With the increase of the bit stream probability of the injected noise, the original edge detection result signal is first enhanced and then submerged. The white noise signal and the original edge detection result signal produce stochastic resonance, the noise energy is transferred to the edge detection result signal, the weak signal is amplified, and the edge detection result image is enhanced, and when the white noise signal is too strong, the original signal will be submerged.

[0087] In another embodiment, as shown, Figure 5 The present disclosure also provides a random computing edge detection enhancement hardware device based on stochastic resonance, which comprises the following parts:

[0088] An image input unit is configured to convert an input image into a gray-scale image, and normalize pixels to obtain a pixel matrix.

[0089] A pulse excitation application unit is configured to apply pulse excitation to the memory.

[0090] A random bit stream generation module is configured to generate a random bit stream by random flipping of the memory under pulse excitation, and the generated random bit stream is used to represent each gray-scale value in the pixel matrix.

[0091] An edge detection random computing unit is configured to perform random computing on the random bit stream representing a specific gray-scale value to obtain an edge detection result.

[0092] A noise capturing unit is configured to capture a noise signal generated when a storage electronic device is working.

[0093] A noise control unit is configured to control and adjust the noise signal to obtain a noise signal representing different random bit streams.

[0094] Although the embodiments of the present application are described above in combination with the drawings, the present application is not limited to the above specific embodiments and application fields, and the above specific embodiments are only illustrative and guiding, but not limiting. Those skilled in the art can make many forms under the guidance of this specification and without departing from the scope protected by the claims of the present application, which are all included in the protection of the present application.

Claims

1. A stochastic computation edge detection enhancement method based on stochastic resonance, characterized in that, Includes the following steps: S100: Acquire an image and use it as an input image, convert the input image into a grayscale image, and normalize the grayscale values ​​to obtain a grayscale value matrix; the normalized grayscale values ​​are decimals between [0,1], and the probability of "1" in the random bit stream is used to represent the grayscale value; S200: A pulse excitation is applied to the memory, causing the memory to randomly flip and generate a random bit stream. This random bit stream represents each gray value in the grayscale matrix. In this step, the pulse excitation applied to the memory is either a current excitation or a voltage excitation, and both the voltage excitation and the current excitation are second-order excitations within one cycle, including a probability setting pulse and a refresh pulse. The probability setting pulse enables the memory to randomly flip from a low-resistance state (P state) to a high-resistance state (AP state) with a certain probability, or from a high-resistance state (AP state) to a low-resistance state (P state) with a certain probability. The refresh pulse enables the memory to flip from its current state to a low-resistance state (P state) or from its current state to a high-resistance state (AP state). The state of the memory is compared with a reference resistance value; a bit "1" is generated in the high-resistance state, and a bit "0" is generated in the low-resistance state. S300: Random calculations are performed on the random bitstream to obtain the edge detection result; edge detection is performed by convolving the template as the kernel with the random bitstream corresponding to the gray value of each pixel in the image, and then a suitable threshold is selected to extract the edge. The template uses the Roberts operator. S400: Captures noise signals generated by electronic devices used for image acquisition during operation, samples and generates a random bit stream with a predetermined probability; by controlling the random flip probability of the memory and the threshold of noise sampling of electronic devices, the noise signal is modulated to achieve the output of a random bit stream with a predetermined probability. S500: Injects a random bitstream of noise with a predetermined probability into the edge detection result to achieve enhancement and optimization.

2. The method for enhancing edge detection based on stochastic resonance according to claim 1, characterized in that, In step S200, the memory includes a magnetic random access memory.

3. The method for enhancing edge detection based on stochastic resonance according to claim 1, characterized in that, In step S200, the memory includes a phase-change memory.

4. The method for enhancing edge detection based on stochastic resonance according to claim 1, characterized in that, In step S200, the memory includes a ferroelectric memory.

5. The method for enhancing edge detection based on stochastic resonance according to claim 1, characterized in that, In step S400, the noise signal includes a random signal generated by the memory.

6. The method for enhancing edge detection based on stochastic resonance according to claim 1, characterized in that, In step S400, the noise signal includes any one or a combination of the following: 1 / f noise, RTN noise, and Gaussian noise.

7. The method for enhancing edge detection based on stochastic resonance according to claim 1, characterized in that, Electronic devices include autonomous vehicles that recognize road markings and obstacles, or medical diagnostic imaging equipment.

8. A system for performing the random computation edge detection enhancement method of claim 1, characterized in that, It includes, The image input unit is used to acquire the input image and convert it into a grayscale image, and to normalize the pixels to obtain a grayscale value matrix. A pulse excitation application unit is used to apply pulse excitation to the memory; The random bit stream generation module is used to generate a random bit stream that represents each gray value in the gray value matrix when the memory is randomly flipped under pulse excitation. The edge detection random computation unit is used to perform random computations on a random bit stream to obtain the edge detection result; A noise capture unit is used to capture noise signals generated by the electronic devices used for image acquisition during operation. The noise control unit is used to control and adjust the noise signal to obtain a random bit stream with a predetermined probability. The random bit stream with the predetermined probability is injected into the edge detection result to achieve enhancement and optimization.

9. A computer storage medium, characterized in that, The storage medium includes computer instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1-7.

10. An electronic device, characterized in that, The electronic device includes: Memory, processor, and computer programs stored in memory and executable on the processor, wherein, When the processor executes the program, it implements the method as described in any one of claims 1-7.

Images