Image exposure detection method, computer readable storage medium and intelligent device

Abstract

This application relates to the field of image processing technology, specifically providing an image exposure detection method, a computer-readable storage medium, and a smart device, aiming to solve the problems of image exposure detection methods, computer-readable storage media, and smart devices. To this end, the image exposure detection method of this application includes: acquiring a synthetic image of a target scene, the synthetic image including segmentation annotation information; determining the contour of a target object in the synthetic image based on the segmentation annotation information, and acquiring the pixel brightness contrast of a preset area around the contour of the target object in the synthetic image; and determining the exposure of the synthetic image based on the pixel brightness contrast. This method, based on pixel brightness contrast, has wide applicability in various scenarios. By acquiring the pixel brightness contrast of a preset area around the contour of the target object, it can effectively reduce the computational load and effectively improve the exposure detection efficiency and accuracy.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and specifically provides an image exposure detection method, a computer-readable storage medium, and an intelligent device. Background Art

[0002] When using an image acquisition device to acquire images of various scenes, it is often easy to have overexposure or underexposure abnormalities, resulting in loss of information in the abnormally exposed area of ​​the image. However, when this is applied to other scenarios such as the image-based target object recognition process, it will affect the recognition results and efficiency. Therefore, in order to avoid the influence of abnormally exposed images, it is usually necessary to effectively detect images with abnormally exposed images.

[0003] In the related art, the following methods are proposed to detect whether an image has abnormal exposure: based on the brightness distribution histogram of pixels in the image, the comparison between the brightness of pixels in the image and the preset brightness threshold, the contrast of the brightness of pixels in the image, the edge information of objects in the image, the color distribution of the image, or deep learning methods, abnormal exposure detection is performed. However, the above methods have problems such as limited application scenarios, large amount of calculation, low detection efficiency, and low accuracy. Summary of the invention

[0004] The present application aims to solve the above-mentioned technical problems, namely, to solve the problems that the existing image exposure anomaly detection methods are limited in application scenarios, have large computational complexity, and have low detection efficiency and accuracy.

[0005] In a first aspect, the present application provides an image exposure detection method, which comprises:

[0006] Acquire a composite image of the target scene, wherein the composite image includes segmentation and annotation information;

[0007] Determine the target object contour of the synthetic image based on the segmentation and annotation information, and obtain the pixel brightness contrast of a preset area around the target object contour in the synthetic image;

[0008] An exposure level of the composite image is determined based on the pixel brightness contrast.

[0009] In some embodiments, acquiring a composite image of a target scene includes:

[0010] Acquire a first image of the target scene at a first exposure level and a second image of the target scene at a second exposure level, wherein the first exposure level is within a normal exposure range of a preset exposure level interval, and the second exposure level is different from the first exposure level;

[0011] Performing image segmentation on the first image and marking each region obtained after the image segmentation to obtain segmentation and marking information of each region in the first image;

[0012] The segmentation annotation information and the second image are synthesized to obtain the synthesized image.

[0013] In some embodiments, obtaining the pixel brightness contrast of a preset area around the outline of the target object in the synthetic image includes:

[0014] Respectively acquiring target pixel brightness of a first preset sub-region and a second preset sub-region around the contour of the target object in the synthetic image, wherein the second preset sub-region and the first preset sub-region do not overlap;

[0015] The pixel brightness contrast is obtained according to a ratio of the target pixel brightness of the first preset sub-area to the target pixel brightness of the second preset sub-area.

[0016] In some embodiments, obtaining the target pixel brightness of a first preset sub-area around the outline of the target object in the synthetic image includes:

[0017] The brightness of target pixels in the synthetic image that are outside the target object outline and are spaced a first distance from the target object outline are obtained.

[0018] In some embodiments, obtaining the target pixel brightness of a second preset sub-area around the outline of the target object in the synthetic image includes:

[0019] The brightness of target pixels in the synthetic image is acquired within the target object contour and within a region spaced a second distance from the target object contour.

[0020] In some embodiments, respectively acquiring the target pixel brightness of the first preset sub-region and the second preset sub-region around the contour of the target object in the synthetic image comprises:

[0021] Acquire first pixel brightness of each pixel in a first preset sub-region around the outline of the target object in the synthetic image, and determine a first average brightness of the first preset sub-region based on the first pixel brightness of each pixel as the target brightness of the first preset sub-region; and

[0022] The second pixel brightness of each pixel in a second preset sub-region around the outline of the target object in the synthetic image is obtained, and the second average brightness of the first preset sub-region is determined based on the second pixel brightness of each pixel as the target brightness of the second preset sub-region.

[0023] In some embodiments, determining the exposure of the composite image based on the pixel brightness contrast comprises:

[0024] Comparing the pixel brightness contrast with a preset threshold;

[0025] When the pixel brightness contrast is less than or equal to the preset threshold, detection determines that the exposure of the composite image is abnormal; when the pixel brightness contrast is greater than the preset threshold, detection determines that the exposure of the composite image is normal.

[0026] In some embodiments, determining the exposure of the composite image based on the pixel brightness contrast comprises:

[0027] Determining whether the pixel brightness contrast is within a preset normal exposure brightness contrast range;

[0028] If not, it is detected that the exposure of the composite image is abnormal; if yes, it is detected that the exposure of the composite image is normal.

[0029] In a second aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the image exposure detection method described in any one of the above items is implemented.

[0030] In a third aspect, the present application provides a smart device, comprising:

[0031] at least one processor;

[0032] and, a memory communicatively coupled to the at least one processor;

[0033] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, any of the above-mentioned image exposure detection methods is implemented.

[0034] In the case of adopting the above technical solution, the present application can obtain a composite image of the target scene, the composite image includes segmentation and annotation information; determine the target object contour of the composite image based on the segmentation and annotation information, and obtain the pixel brightness contrast of a preset area around the target object contour in the composite image; determine the exposure of the composite image based on the pixel brightness contrast. This method is applicable to a wide range of scenarios based on pixel brightness contrast. By obtaining the pixel brightness contrast of a preset area around the target object contour, the amount of calculation can be effectively reduced, and the efficiency of exposure detection can be effectively improved. At the same time, by performing exposure detection based on the local area around the target object contour, the detection accuracy can be effectively improved, avoiding the problem that the characteristics of the local area cannot be effectively reflected when detecting based on the entire image, affecting the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, in which:

[0036] Figure 1It is a flowchart of an image exposure detection method provided in an embodiment of the present application;

[0037] Figure 2 It is a schematic flow chart of a method for obtaining a composite image of a target scene provided in an embodiment of the present application;

[0038] Figure 3 It is a flowchart of a method for obtaining the pixel brightness contrast of a preset area around the outline of a target object in a synthetic image provided by an embodiment of the present application;

[0039] Figure 4 is a composite image of a garage entrance and exit scene provided by this application;

[0040] Figure 5 It is a schematic diagram of the structure of the smart device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0041] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the protection scope of the present application.

[0042] Based on the description of the background technology section, it can be seen that the exposure detection methods provided in the related art have problems such as limited applicable scenarios, large amount of calculation, low detection efficiency and low accuracy. For example, the method based on the histogram of the brightness distribution of pixels in the image performs histogram analysis based on the entire image, which has limited effect on complex scenes and images with uneven grayscale distribution, and low detection accuracy. The method based on the comparison of the brightness of pixels in the image and the preset brightness threshold usually requires manual setting of the threshold, which is not suitable for dynamically changing environments or lighting conditions, and the preset brightness threshold is usually set for the entire image, which cannot effectively reflect the situation of the local area, and the detection accuracy for images with uneven lighting is low. The method based on the contrast of the brightness of image pixels has a large amount of calculation and may not be efficient enough for large images or real-time systems. The method based on the edge information of objects in the image has poor detection effect on images with blurred edges and limited applicable scenarios. The method based on the color distribution of the image has low accuracy for grayscale images or images with uneven lighting and limited applicable scenarios. The method based on deep learning requires a large amount of labeled data for training, has high requirements for computing resources and algorithm tuning, and requires network structure design and parameter adjustment to adapt to different tasks, which is not convenient enough.

[0043] In view of this, the present application provides an image exposure detection method based on the fact that whether the target object is visible is directly related to whether the target object outline is clearly visible, and whether the exposure is abnormal will affect whether the target is visible. The target object outline is determined by using the segmentation and annotation information of the synthetic image, and whether the target object is visible is determined based on the pixel brightness contrast of the preset area around the target object outline, so as to determine the exposure of the synthetic image. While effectively reducing the amount of calculation, it can effectively improve the detection efficiency and accuracy and has a wide range of applicable scenarios.

[0044] See also Figure 1 As shown, Figure 1 : is a flow chart of an image exposure detection method provided in an embodiment of the present application, which may include:

[0045] Step S11: obtaining a composite image of the target scene, where the composite image includes segmentation and annotation information;

[0046] Step S12: determining the target object contour of the synthesized image based on the segmentation and annotation information, and obtaining the pixel brightness contrast of a preset area around the target object contour in the synthesized image;

[0047] Step S13: Determine the exposure of the composite image based on the pixel brightness contrast.

[0048] The image exposure detection method provided in this application can be applied to a variety of target scenes such as driving scenes, garage entrances and exits, etc.

[0049] The composite image may be obtained by synthesizing multiple different images of the target scene using fusion, splicing and other techniques, wherein at least some of the multiple different images are annotated with segmentation annotation information, and during the synthesis process, the segmentation annotation information of the partial images can be correctly matched to the corresponding position in the composite image, thereby obtaining a composite image with segmentation annotation information. In some embodiments, the segmentation annotation information can be used to characterize the category or object to which each pixel or region in the image belongs.

[0050] In some embodiments, see Figure 2 As shown, step S11 may specifically be:

[0051] Step S111: acquiring a first image of a target scene at a first exposure level and a second image of a target scene at a second exposure level, wherein the first exposure level is within a preset exposure level range, and the second exposure level is different from the first exposure level;

[0052] Step S112: performing image segmentation on the first image and marking each region obtained after the image segmentation to obtain segmentation and marking information of each region in the first image;

[0053] Step S113: synthesize the segmentation annotation information and the second image to obtain a synthesized image.

[0054] In an embodiment of the present application, the preset exposure interval can be a range corresponding to a normal exposure, which can be flexibly set according to the actual application scenario. Accordingly, the first image obtained at the first exposure is an image obtained under normal exposure conditions, and the contours of each object in the image are clear, which facilitates image segmentation.

[0055] In some embodiments, step S111 may specifically include: obtaining first images of the target scene at multiple different first exposures and second images at multiple different second exposures, wherein the multiple different first exposures are all within a preset exposure range, and the second exposure is different from the first exposure.

[0056] The second exposure level may be within a preset exposure range or may not be within the preset exposure range. When the second exposure level does not belong to the preset exposure range, the corresponding second image may be an image that is prone to exposure abnormalities.

[0057] In some embodiments, step S112 may specifically be to segment the first image using an image segmentation technology and to label each region obtained after the image segmentation, so as to obtain segmentation and labeling information of each region in the first image.

[0058] In the embodiment of the present application, the composite image may include multiple objects of the target scene, and accordingly, at least one object may be acquired from the composite image as the target object and its contour may be acquired as the target object contour.

[0059] In some embodiments, determining the target object contour of the synthesized image based on the segmentation and annotation information in step S12 may specifically be determining the target object contour of the synthesized image based on the segmentation and annotation information using an edge detection algorithm.

[0060] In some embodiments, the preset area may include a first preset sub-area and a second preset sub-area, see Figure 3 As shown, the pixel brightness contrast of the preset area around the outline of the target object in the synthetic image obtained in step S12 can be specifically:

[0061] Step S31: respectively obtaining target pixel brightness of a first preset sub-region and a second preset sub-region around the contour of the target object in the synthesized image, wherein the second preset sub-region does not overlap with the first preset sub-region;

[0062] Step S32: Obtaining a pixel brightness contrast ratio according to a ratio of a target pixel brightness of the first preset sub-region to a target pixel brightness of the second preset sub-region.

[0063] The first preset sub-region and the second preset sub-region may correspond to the area inside and outside the outline divided by the outline of the target object, respectively. In some embodiments, step S31 may be specifically as follows:

[0064] Acquire the brightness of target pixels in a region outside the outline of the target object and spaced a first distance from the outline of the target object in the synthesized image;

[0065] The brightness of target pixels in the synthesized image within the contour of the target object and within a region spaced a second distance from the contour of the target object is obtained.

[0066] The first distance and the second distance can be flexibly set as needed.

[0067] In some embodiments, the target pixel brightness may be the average brightness of the pixels in the corresponding preset sub-area, and step S31 may be specifically as follows:

[0068] Acquire the first pixel brightness of each pixel in a first preset sub-region around the outline of the target object in the synthetic image, and determine a first average brightness of the first preset sub-region based on the first pixel brightness of each pixel as the target brightness of the first preset sub-region; and

[0069] The second pixel brightness of each pixel in the second preset sub-region around the contour of the target object in the synthetic image is obtained, and the second average brightness of the first preset sub-region is determined based on the second pixel brightness of each pixel as the target brightness of the second preset sub-region.

[0070] See also Figure 4 As shown, Figure 4 It is a composite image of the garage entrance and exit scene provided in the present application, in which the outline of the target object shown is a cone outline, the first preset sub-area can be the area between the cone outline and the triangle shown inside the cone outline, and the second preset sub-area can be the area between the cone outline and the triangle shown outside the cone outline.

[0071] When the exposure is abnormal, such as overexposure or underexposure, the pixel brightness of the preset area around the outline of the target object in the composite image is closer. In some embodiments, step S13 can be specifically as follows:

[0072] Compare pixel brightness contrast with a preset threshold;

[0073] When the pixel brightness contrast is less than or equal to a preset threshold, it is determined that the exposure of the composite image is abnormal; when the pixel brightness contrast is greater than the preset threshold, it is determined that the exposure of the composite image is normal.

[0074] Among them, the preset threshold can be set according to the actual application scenario, and can be set according to the local area of ​​the image, which is conducive to improving the accuracy of detection.

[0075] In some other embodiments, step S13 may specifically be:

[0076] Determine whether the pixel brightness contrast is within a preset exposure brightness contrast range;

[0077] If not, it is determined that the exposure of the composite image is abnormal; if yes, it is determined that the exposure of the composite image is normal.

[0078] The preset exposure brightness contrast range can be set according to actual needs.

[0079] The above is the image exposure detection method provided by the embodiment of the present application, which obtains a composite image of the target scene, wherein the composite image includes segmentation and annotation information; determines the target object contour of the composite image based on the segmentation and annotation information, and obtains the pixel brightness contrast of a preset area around the target object contour in the composite image; and determines the exposure of the composite image based on the pixel brightness contrast. This method is applicable to a wide range of scenarios based on pixel brightness contrast. By obtaining the pixel brightness contrast of a preset area around the target object contour, the amount of calculation can be effectively reduced, and the efficiency of exposure detection can be effectively improved. At the same time, by performing exposure detection based on the local area around the target object contour, the detection accuracy can be effectively improved, avoiding the problem that the characteristics of the local area cannot be effectively reflected when the detection is based on the entire image, which affects the detection accuracy.

[0080] It is understood by those skilled in the art that all or part of the processes in the method for implementing the above-mentioned embodiments of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0081] Another aspect of the present application further provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by a processor, the image exposure detection method in any of the above embodiments can be implemented. The computer-readable storage medium can be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-transitory computer-readable storage medium.

[0082] In another aspect of the present application, a smart device is provided.

[0083] In an embodiment of an intelligent device according to the present application, the intelligent device may include at least one processor; and a memory connected to the at least one processor in communication; wherein a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method described in any of the above embodiments is implemented. The intelligent device described in the present application may include a driving device, a smart car, a robot, and the like. Figure 5 , Figure 5 FIG. 5 exemplarily shows that the memory 51 and the processor 52 are communicatively connected via a bus.

[0084] In some embodiments of the present application, the smart device may further include at least one sensor for sensing information. The sensor is communicatively connected to any type of processor mentioned in the present application. Optionally, the smart device may further include an autonomous driving system for guiding the smart device to drive itself or assist in driving. The processor communicates with the sensor and / or the autonomous driving system to complete the method described in any of the above embodiments.

[0085] The relevant user personal information that may be involved in the various embodiments of this application is strictly in accordance with the requirements of laws and regulations, following the principles of legality, legitimacy and necessity, based on the reasonable purposes of business scenarios, to process the personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as the personal information obtained with the user's authorization.

[0086] The user personal information processed by this application will vary depending on the specific product / service scenario, and shall be based on the specific scenario in which the user uses the product / service, and may involve the user's account information, device information, driving information, vehicle information or other related information. This application will treat the user's personal information and its processing with a high degree of diligence.

[0087] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that meet industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.

[0088] So far, the technical solutions of the present application have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present application.

Claims

1. A method for detecting image exposure, characterized in that: include: Acquire a composite image of the target scene, wherein the composite image includes segmentation and annotation information; Determining the target object contour of the synthetic image based on the segmentation and annotation information, and obtaining the pixel brightness contrast of a preset area around the target object contour in the synthetic image, including: respectively obtaining the target pixel brightness of a first preset sub-area and a second preset sub-area around the target object contour in the synthetic image, wherein the second preset sub-area and the first preset sub-area do not overlap; obtaining the pixel brightness contrast according to the ratio of the target pixel brightness of the first preset sub-area to the target pixel brightness of the second preset sub-area; wherein the first preset sub-area is a local area outside the target object contour and spaced a first distance from the target object contour, and the second preset sub-area is a local area inside the target object contour and spaced a second distance from the target object contour; Determining the exposure of the synthetic image based on the pixel brightness contrast; wherein acquiring the synthetic image of the target scene comprises: Acquire a first image of the target scene at a first exposure and a second image at a second exposure, wherein the first exposure is within a preset exposure range, and the second exposure is different from the first exposure; the preset exposure range is a range corresponding to a normal exposure; Performing image segmentation on the first image and marking each region obtained after the image segmentation to obtain segmentation and marking information of each region in the first image; The segmentation annotation information and the second image are synthesized to obtain the synthesized image.

2. The method according to claim 1, characterized in that The step of respectively acquiring target pixel brightness of a first preset sub-region and a second preset sub-region around the contour of the target object in the synthetic image comprises: Acquire first pixel brightness of each pixel in a first preset sub-region around the outline of the target object in the synthetic image, and determine a first average brightness of the first preset sub-region based on the first pixel brightness of each pixel as the target brightness of the first preset sub-region; and The second pixel brightness of each pixel in a second preset sub-region around the outline of the target object in the synthetic image is obtained, and the second average brightness of the first preset sub-region is determined based on the second pixel brightness of each pixel as the target brightness of the second preset sub-region.

3. The method according to claim 1, characterized in that The determining the exposure of the composite image based on the pixel brightness contrast comprises: Comparing the pixel brightness contrast with a preset threshold; When the pixel brightness contrast is less than or equal to the preset threshold, it is determined that the exposure of the composite image is abnormal; when the pixel brightness contrast is greater than the preset threshold, it is determined that the exposure of the composite image is normal.

4. The method according to claim 1, characterized in that: The determining the exposure of the composite image based on the pixel brightness contrast comprises: Determining whether the pixel brightness contrast is within a preset exposure brightness contrast range; If not, it is determined that the exposure of the composite image is abnormal; if so, it is determined that the exposure of the composite image is normal.

5. 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 image exposure detection method according to any one of claims 1 to 4 is implemented.

6. A smart device, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the image exposure detection method according to any one of claims 1 to 4 is implemented.

Images