A method, device, equipment and medium for obtaining a salient region

Abstract

The present application discloses a method, device, equipment and medium for acquiring a salient area, which is used to solve the technical problem that the existing computer vision cannot clearly perceive the key areas and other areas in the image, and thus cannot acquire the key areas in time. The method includes: acquiring a single-frame video image and performing filtering processing; performing Gaussian pyramid processing on the filtered single-frame video image to obtain intensity features, and obtaining an intensity feature map through center-surround difference operation and cross-scale summation operation; performing Gaussian pyramid processing and filtering processing on the filtered single-frame video image to obtain texture features in four directions, and obtaining a corresponding texture feature map through center-surround difference operation and cross-scale summation operation; linearly fusing the intensity feature map with the texture feature map to obtain the corresponding saliency map, and performing segmentation processing and morphological processing to obtain the salient area corresponding to the single-frame video image, thereby improving the contrast of the salient area.

Description

Technical Field

[0001] The present application relates to the field of computer vision perception technology, and in particular to a method, device, equipment and medium for acquiring a salient area. Background Art

[0002] When humans observe natural scenes, they often only focus on part of the scene, or consciously observe only a specific scene in the scene. This area or specific area that is focused on is called a salient area. The purpose of salient area extraction is to locate the most sensitive and eye-catching area in the image.

[0003] Computer vision often needs to first perceive the key areas in the field of vision, which helps the subject take corresponding measures for the first perceived area. Research on visual information processing mechanisms has become a hot topic in the field of computer vision. The extraction of salient areas is very useful and can be used for accurate positioning of a series of subsequent operations such as image segmentation, recognition, compression, and encoding. Summary of the invention

[0004] The embodiments of the present application provide a method, device, equipment and medium for acquiring a salient area, so as to solve the technical problem that the existing computer vision cannot clearly perceive the key areas and other areas in the image, and thus cannot acquire the key areas in time.

[0005] On the one hand, an embodiment of the present application provides a method for obtaining a salient region, comprising:

[0006] Acquire a single-frame video image from a video sequence, and perform filtering processing on the single-frame video image;

[0007] Performing Gaussian pyramid processing on the filtered single-frame video image to obtain intensity features, and obtaining a corresponding intensity feature map through center-surrounding difference calculation and cross-scale summation calculation;

[0008] and performing the Gaussian pyramid processing and filtering processing on the filtered single-frame video image to obtain texture features in four directions, and obtaining a texture feature map corresponding to each direction through the center-around difference operation and the cross-scale summation operation;

[0009] The texture feature maps corresponding to the four directions are summed to obtain a texture feature salient map corresponding to the single-frame video image, and the texture feature salient map corresponding to the single-frame video image is linearly fused with the intensity feature map to obtain a corresponding saliency map;

[0010] The saliency map is segmented and morphologically processed to obtain a saliency region corresponding to the single-frame video image.

[0011] In one implementation of the present application, the Gaussian pyramid processing is performed on the filtered single-frame video image to obtain intensity features, and a corresponding intensity feature map is obtained by center-surround difference operation and cross-scale summation operation, specifically including:

[0012] Based on the Gaussian pyramid, the filtered single-frame video image is divided into four first images of different sizes, and corresponding intensity features are obtained;

[0013] Perform center-periphery difference operations on four first images of different sizes in pairs, so as to obtain six corresponding first fused images;

[0014] A cross-scale summation operation is performed on the six corresponding first fused images to obtain a corresponding intensity feature map.

[0015] In one implementation of the present application, the Gaussian pyramid processing and filtering processing are performed on the filtered single-frame video image to obtain texture features in four directions, and a texture feature map corresponding to each direction is obtained by performing the center-around difference operation and the cross-scale summation operation, specifically including:

[0016] Based on the Gaussian pyramid, dividing the filtered single-frame video image into three second images of different sizes;

[0017] Select any Gabor filter from four Gabor filters in different directions, and perform convolution operations on the three second images of different sizes with the corresponding current Gabor filters of different scales, respectively, to obtain three third images of different sizes;

[0018] Perform the center-periphery difference operation on the three third images of different sizes in pairs, so as to obtain three second fused images corresponding to the current direction;

[0019] Performing the cross-scale summation operation on the corresponding three second fused images to obtain a texture feature map in the current direction;

[0020] The Gabor filters in the other three directions except the current Gabor filter are traversed to obtain texture feature maps corresponding to the other three directions.

[0021] In one implementation of the present application, it also includes:

[0022] When performing the center-surrounding difference calculation, according to the following formula:

[0023]

[0024] Among them, I(c,s) represents the image obtained by the center-surrounding difference operation, I(c) represents the center image, and I(s) represents the surrounding image. is the cross-scale operation symbol of the center image and the surrounding images.

[0025] In one implementation of the present application, it also includes:

[0026] When performing the cross-scale summation operation, according to the following formula:

[0027]

[0028] in, represents the feature salient map, is the cross-scale summation operator, N represents the normalization operator, I(c,s) represents the image obtained by the center-surrounding difference operation, c represents the center, and s represents the surrounding.

[0029] In one implementation of the present application, it also includes:

[0030] When the texture feature maps corresponding to the four directions are summed, according to the following formula:

[0031]

[0032] in, represents the fusion texture feature salient map, θ represents the direction of the filter, θ = {0°, 45°, 90°, 135°}, is the cross-scale summation operator symbol, and N represents the normalization operator symbol.

[0033] In one implementation of the present application, the filtering process on the single-frame video image specifically includes:

[0034] The single-frame video image is filtered by median filtering and homomorphic filtering to eliminate noise in the single-frame video image and enhance the image.

[0035] On the other hand, an embodiment of the present application further provides a device for acquiring a salient region, the device comprising:

[0036] An image acquisition unit, used for acquiring a single-frame video image from a video sequence, and performing filtering processing on the single-frame video image;

[0037] An intensity feature acquisition unit is used to perform Gaussian pyramid processing on the filtered single-frame video image to obtain an intensity feature, and obtain a corresponding intensity feature map through center-surrounding difference calculation and cross-scale summation calculation;

[0038] A texture feature acquisition unit, configured to perform the Gaussian pyramid processing and filtering processing on the filtered single-frame video image to obtain texture features in four directions, and obtain a texture feature map corresponding to each direction through the center-surrounding difference operation and the cross-scale summation operation;

[0039] A linear fusion unit, used for summing the texture feature maps in four directions to obtain a texture feature salient map corresponding to the single-frame video image, and linearly fusing the texture feature salient map corresponding to the single-frame video image with the intensity feature map to obtain a corresponding saliency map;

[0040] The salient region acquisition unit is used to perform segmentation processing and morphological processing on the salient map to obtain the salient region corresponding to the single-frame video image.

[0041] On the other hand, an embodiment of the present application further provides a device for acquiring a salient region, the device comprising:

[0042] at least one processor;

[0043] and, a memory communicatively coupled to the at least one processor;

[0044] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0045] Execute any one of the above methods for acquiring a salient region.

[0046] On the other hand, an embodiment of the present application further provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured as follows:

[0047] A method for obtaining a salient region as described in any of the above items.

[0048] The embodiments of the present application provide a method, device, equipment and medium for acquiring a salient region, which at least include the following beneficial effects: enhancing an image by filtering a single-frame video image, processing the filtered single-frame video image by Gaussian pyramid to obtain a corresponding intensity feature, and obtaining an intensity feature map corresponding to the single-frame video image by center-surround difference operation and cross-scale summation operation; obtaining texture features in four directions by performing Gaussian pyramid operation and filtering operation on the filtered single-frame video image; and obtaining a texture feature map corresponding to each direction by center-surround difference operation and cross-scale operation; obtaining a texture feature highlight map corresponding to the single-frame video image by summing the texture feature maps in four directions, and then linearly fusing the texture feature highlight map corresponding to the single-frame video image and the intensity feature map to obtain a saliency map corresponding to the single-frame video image, and finally obtaining a saliency region corresponding to the single-frame video image by segmentation and morphological processing on the saliency map, thereby improving the contrast between the saliency region and other regions in the image, and being able to highlight the saliency region, so as to facilitate accurate positioning of the target in subsequent use. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0050] Figure 1 A schematic diagram of a flow chart of a method for acquiring a salient region provided in an embodiment of the present application;

[0051] Figure 2 A schematic diagram of the structure of a device for acquiring a salient region provided in an embodiment of the present application;

[0052] Figure 3 A schematic diagram of the internal structure of a salient region acquisition device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. 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 making creative work are within the scope of protection of the present application.

[0054] The embodiment of the present application provides a method, device, equipment and medium for acquiring a salient region, which enhances the image by filtering a single-frame video image, processes the filtered single-frame video image by Gaussian pyramid, obtains the corresponding intensity feature, obtains the intensity feature map corresponding to the single-frame video image by center-surround difference operation and cross-scale summation operation; obtains the texture features in four directions by Gaussian pyramid operation and filtering operation on the filtered single-frame video image; obtains the texture feature map corresponding to each direction by center-surround difference operation and cross-scale operation; obtains the texture feature highlight map corresponding to the single-frame video image by summing the texture feature maps in four directions, and then linearly fuses the texture feature highlight map corresponding to the single-frame video image and the intensity feature map, so as to obtain the salient region corresponding to the single-frame video image, and finally obtains the salient region corresponding to the single-frame video image by segmentation and morphological processing on the salient region. The technical problem that the existing computer vision cannot clearly perceive the key areas and other areas in the image, and thus cannot obtain the key areas in time is solved.

[0055] The technical solution proposed in the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0056] Figure 1 The following is a flow chart of a method for obtaining a salient region provided in an embodiment of the present application. Figure 1 As shown, a method for obtaining a salient region provided in an embodiment of the present application mainly includes the following steps:

[0057] S101, obtaining a single-frame video image from a video sequence, and performing filtering processing on the single-frame video image.

[0058] The server obtains a single-frame video image from the monitored video sequence, and uses the single-frame video image as the original image, and performs filtering processing on the image, thereby enhancing the image.

[0059] Specifically, the server performs filtering processing on the single-frame video image through median filtering and homomorphic filtering, thereby eliminating noise in the single-frame video image, and can also enhance the single-frame video image to facilitate subsequent operations on the denoised and enhanced single-frame video image.

[0060] S102, performing Gaussian pyramid processing on the filtered single-frame video image to obtain intensity features, and obtaining a corresponding intensity feature map through center-surrounding difference operation and cross-scale summation operation.

[0061] The server performs Gaussian pyramid processing on the filtered single-frame video image to obtain the intensity features corresponding to the single-frame video image, and then obtains the intensity feature map corresponding to the single-frame video image through center-surround difference operation and cross-scale summation operation.

[0062] Gaussian pyramid is a technology used in image processing, computer vision, and signal processing. When processing an image, the Gaussian pyramid takes the current image as the original image and performs subsequent processing operations on the basis of the original image. Gaussian pyramid is essentially a multi-scale representation of signals. It is a simple and effective structure for realizing multi-scale analysis of images. The same signal or image is Gaussian blurred multiple times and downsampled to produce multiple groups of signals or images at different sizes. An image pyramid is a set of images arranged in a pyramid form with gradually decreasing resolutions. The bottom of the pyramid image is represented by the high resolution of the original image, and the top is represented by the low resolution of the image. It should be noted that sampling in the direction of decreasing scale is called downsampling, and sampling in the direction of increasing scale is called upsampling. Gaussian pyramid is achieved by Gaussian blurring and downsampling the original image layer by layer from 1:1 to 1:8.

[0063] Specifically, based on the Gaussian pyramid, the server divides the filtered single-frame video image into four layers, that is, into four first images of different sizes, and obtains the intensity feature corresponding to the single-frame video image. The server then performs center-periphery difference operations on the four first images of different sizes, thereby obtaining six corresponding first fused images, and then performs cross-scale summation operations on the six first fused images, so as to obtain the intensity feature map corresponding to the single-frame video image, so as to facilitate subsequent operations based on the intensity feature map corresponding to the single-frame video image.

[0064] In one embodiment of the present application, the server calculates the center-around difference operation according to the following formula:

[0065]

[0066] It should be noted that I(c, s) in the embodiment of the present application represents an image obtained by performing a center-surrounding difference operation, I(c) represents a center image, and I(s) represents a surrounding image. is the cross-scale operation symbol of the center image and the surrounding images.

[0067] In one embodiment of the present application, when the server performs the cross-scale summation operation, it is performed according to the following formula:

[0068]

[0069] It should be noted that the embodiments of the present application represents the feature salient map, is the cross-scale summation operator, N represents the normalization operator, I(c,s) represents the image obtained by the center-surrounding difference operation, c represents the center, and s represents the surrounding.

[0070] S103, performing Gaussian pyramid processing and filtering processing on the filtered single-frame video image to obtain texture features in four directions, and obtaining a texture feature map corresponding to each direction through center-around difference operation and cross-scale summation operation.

[0071] The server performs Gaussian pyramid processing and filtering on the filtered single-frame video image to obtain texture features in four directions, and then obtains the texture feature map corresponding to each direction through center-surrounding difference operation and cross-scale summation operation.

[0072] Gabor wavelet is a wavelet that is similar to the visual stimulus response of simple cells in the human visual system. Gabor wavelet is sensitive to image edges, can provide good direction selection and scale selection characteristics, and is insensitive to changes in illumination. This application uses the Gabor kernel function as a filter. The frequency and direction representation of the Gabor filter is close to the frequency and direction of the human visual system. Gabor image processing is to perform convolution operations with the image using the Gabor filter. The Gabor filter here refers to a 2D Gabor filter.

[0073] Specifically, based on the Gaussian pyramid, the server divides the filtered single-frame video image into three second images of different sizes, selects any Gabor filter from the Gabor filters in four different directions, and performs convolution operations on the three second images of different sizes with the corresponding current Gabor filters of different scales to obtain three third images of different sizes; performs center-periphery difference operations on the three third images of different sizes in pairs, so as to obtain three second fused images corresponding to the current direction; and performs cross-scale summation operations on the corresponding three second fused images to obtain a texture feature map in the current direction.

[0074] Traverse the Gabor filters in the other three directions except the current Gabor filter, and based on the same operation as above, obtain the texture feature maps corresponding to the other three directions.

[0075] S104, summing the texture feature maps in four directions to obtain a texture feature map corresponding to a single-frame video image, and linearly fusing the texture feature map corresponding to the single-frame video image with the intensity feature map to obtain a corresponding saliency map.

[0076] The server can obtain a direction-insensitive texture feature map corresponding to a single-frame video image by summing up the texture feature maps in four directions, and then linearly fuses the texture feature map and intensity feature map corresponding to the acquired single-frame video image to obtain a fused saliency map.

[0077] Specifically, when the server sums the corresponding texture feature maps in the four directions, it is performed according to the following formula:

[0078]

[0079] It should be noted that the embodiments of the present application represents the fusion texture feature salient map, θ represents the direction of the filter, and θ in this application = {0°, 45°, 90°, 135°}, is the cross-scale summation operator symbol, and N represents the normalization operator symbol.

[0080] In one embodiment of the present application, a large-scale image is defined as a fine image, and a small-scale image is defined as a coarse image. By performing a bicubic interpolation operation from the coarse image to the fine image, a group of images in a cross-scale subtraction operation are made to have the same size.

[0081] S105: Perform segmentation processing and morphological processing on the saliency map to obtain a salient region corresponding to a single-frame video image.

[0082] After obtaining the saliency map corresponding to a single-frame video image, the server performs segmentation and morphological processing on the saliency map, thereby obtaining the connected or discrete saliency areas corresponding to the single-frame video image, and then performs operations on it to obtain the maximum connected area.

[0083] The above is an embodiment of the method proposed in this application. Based on the same inventive concept, this application embodiment also provides a device for obtaining a salient region, and its structure is as follows: Figure 2 shown.

[0084] Figure 2 The schematic diagram of the structure of a device for acquiring a salient region provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the device includes: an image acquisition unit 201, an intensity feature acquisition unit 202, a texture feature acquisition unit 203, a linear fusion unit 204, and a salient region acquisition unit 205.

[0085] The image acquisition unit 201 is used to acquire a single-frame video image from a video sequence and perform filtering processing on the single-frame video image;

[0086] The intensity feature acquisition unit 202 is used to perform Gaussian pyramid processing on the filtered single-frame video image to obtain intensity features, and obtain a corresponding intensity feature map through center-surrounding difference calculation and cross-scale summation calculation;

[0087] The texture feature acquisition unit 203 is used to perform Gaussian pyramid processing and filtering processing on the filtered single-frame video image to obtain texture features in four directions, and obtain a texture feature map corresponding to each direction through center-around difference calculation and cross-scale summation calculation;

[0088] The linear fusion unit 204 is used to sum the texture feature maps in four directions to obtain the texture feature map corresponding to the single-frame video image, and linearly fuse the texture feature map corresponding to the single-frame video image with the intensity feature map to obtain the corresponding saliency map;

[0089] The salient region acquisition unit 205 is used to perform segmentation processing and morphological processing on the salient map to obtain the salient region corresponding to the single-frame video image.

[0090] Figure 3 The internal structure diagram of a salient region acquisition device provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the device includes:

[0091] at least one processor;

[0092] and, a memory communicatively coupled to the at least one processor;

[0093] The memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor to enable the at least one processor to:

[0094] Execute any one of the above methods for acquiring a salient region.

[0095] The embodiment of the present application further provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured as follows:

[0096] A method for obtaining a salient region as described in any of the above items.

[0097] In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0098] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0099] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0100] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0101] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0102] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0105] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0106] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0107] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0108] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0109] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0110] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0111] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0112] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of this specification.

Claims

1. A method for obtaining a salient region, characterized in that: The method comprises: Acquire a single-frame video image from a video sequence, and perform filtering processing on the single-frame video image; Performing Gaussian pyramid processing on the filtered single-frame video image to obtain intensity features, and obtaining a corresponding intensity feature map through center-surrounding difference calculation and cross-scale summation calculation; Performing the Gaussian pyramid processing and filtering processing on the filtered single-frame video image to obtain texture features in four directions, and obtaining a texture feature map corresponding to each direction through the center-around difference operation and the cross-scale summation operation; The texture feature maps in four directions are summed to obtain a texture feature salient map corresponding to the single-frame video image, and the texture feature salient map corresponding to the single-frame video image is linearly fused with the intensity feature map to obtain a corresponding saliency map; The saliency map is segmented and morphologically processed to obtain a saliency region corresponding to the single-frame video image.

2. The method for obtaining a salient region according to claim 1, characterized in that: The step of performing Gaussian pyramid processing on the filtered single-frame video image to obtain intensity features, and obtaining a corresponding intensity feature map through center-surrounding difference calculation and cross-scale summation calculation, specifically includes: Based on the Gaussian pyramid, the filtered single-frame video image is divided into four first images of different sizes, and corresponding intensity features are obtained; Perform center-periphery difference operations on four first images of different sizes in pairs, so as to obtain six corresponding first fused images; A cross-scale summation operation is performed on the six corresponding first fused images to obtain a corresponding intensity feature map.

3. The method for obtaining a salient region according to claim 1, characterized in that: The Gaussian pyramid processing and filtering processing are performed on the filtered single-frame video image to obtain texture features in four directions, and a texture feature map corresponding to each direction is obtained by performing the center-around difference operation and the cross-scale summation operation, specifically including: Based on the Gaussian pyramid, dividing the filtered single-frame video image into three second images of different sizes; Select any Gabor filter from four Gabor filters in different directions, and perform convolution operations on the three second images of different sizes with the corresponding current Gabor filters of different scales, respectively, to obtain three third images of different sizes; Perform the center-periphery difference operation on the three third images of different sizes in pairs, so as to obtain three second fused images corresponding to the current direction; Performing the cross-scale summation operation on the corresponding three second fused images to obtain a texture feature map in the current direction; The Gabor filters in the other three directions except the current Gabor filter are traversed to obtain texture feature maps corresponding to the other three directions.

4. The method for obtaining a salient region according to claim 1, characterized in that: The method further comprises: When performing the center-surrounding difference calculation, according to the following formula: Among them, I(c,s) represents the image obtained by the center-surrounding difference operation, I(c) represents the center image, and I(s) represents the surrounding image. is the cross-scale operation symbol of the center image and the surrounding images.

5. The method for obtaining a salient region according to claim 1, characterized in that: The method further comprises: When performing the cross-scale summation operation, according to the following formula: in, represents the feature salient map, is the cross-scale summation operator, N represents the normalization operator, I(c,s) represents the image obtained by the center-surrounding difference operation, c represents the center, and s represents the surrounding.

6. The method for obtaining a salient region according to claim 1, characterized in that: The method further comprises: When the texture feature maps corresponding to the four directions are summed, according to the following formula: in, represents the fusion texture feature salient map, θ represents the direction of the filter, θ = {0°, 45°, 90°, 135°}, is the cross-scale summation operator symbol, and N represents the normalization operator symbol.

7. The method for obtaining a salient region according to claim 1, characterized in that: The filtering process of the single frame video image specifically includes: The single-frame video image is filtered by median filtering and homomorphic filtering to eliminate noise in the single-frame video image and enhance the image.

8. A device for acquiring a salient region, characterized in that: The device comprises: An image acquisition unit, used for acquiring a single-frame video image from a video sequence, and performing filtering processing on the single-frame video image; An intensity feature acquisition unit is used to perform Gaussian pyramid processing on the filtered single-frame video image to obtain an intensity feature, and obtain a corresponding intensity feature map through center-surrounding difference calculation and cross-scale summation calculation; A texture feature acquisition unit, configured to perform the Gaussian pyramid processing and filtering processing on the filtered single-frame video image to obtain texture features in four directions, and obtain a texture feature map corresponding to each direction through the center-surrounding difference operation and the cross-scale summation operation; A linear fusion unit, used for summing the texture feature maps in four directions to obtain a texture feature salient map corresponding to the single-frame video image, and linearly fusing the texture feature salient map corresponding to the single-frame video image with the intensity feature map to obtain a corresponding saliency map; The salient region acquisition unit is used to perform segmentation processing and morphological processing on the salient map to obtain the salient region corresponding to the single-frame video image.

9. A device for acquiring a salient region, characterized in that: The device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: Execute the method for acquiring a salient area as described in any one of claims 1 to 7.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to: Execute the method for acquiring a salient area as described in any one of claims 1 to 7.

Images