Method and apparatus for detecting backlighting of an image

Abstract

Disclosed are a method and apparatus for detecting backlight of an image. An apparatus for detecting backlight of an image includes a processor configured to convert pixel values of each frame of a plurality of image frames based on a first threshold value, block the pixel values into a plurality of block regions to generate a converted frame, generate brightness accumulation data based on the converted frame, determine a block region having a maximum block value in the brightness accumulation data as a backlight candidate block region, classify the block regions into first and second regions based on a position of the backlight candidate block region, and calculate a backlight score using a sum of the block values of the block regions of the first region and a sum of the block values of the block regions of the second region, and determine the image as a backlight image based on a second threshold value and the backlight score.

Description

Technical Field

[0001] This disclosure relates to backlight detection, and more specifically, to a method and apparatus for detecting backlight in images produced by sunlight. Background Technology

[0002] The statements in this section provide only background information in relation to this disclosure and do not necessarily constitute prior art.

[0003] In autonomous driving and advanced driver assistance systems (ADAS), vehicles use various sensors (such as cameras, radar, lidar, and ultrasonic sensors) to identify the surrounding environment and driving conditions. Among these, images captured by cameras can provide the vehicle with visual information about the surrounding environment, such as human visual information.

[0004] With the development of autonomous driving technology, various image recognition systems using cameras to identify the surrounding environment have been applied to vehicles. However, due to the various light sources present around the vehicle, backlighting may occur, and due to brightness distortion in some areas, it is difficult to accurately interpret backlit images taken under backlight conditions. In particular, backlighting caused by strong sunlight may lead to recognition errors or failures in the image recognition system, which increases the risk of accidents involving autonomous vehicles and reduces the reliability of autonomous driving.

[0005] To address the problem of backlighting in images, numerous methods have been proposed for detecting and correcting backlighting from still images or moving images. However, because existing methods are designed to uniformly determine the presence of light without considering the type and intensity of the light source, backlighting correction can be performed even in weak backlighting conditions where image recognition is possible, potentially leading to unnecessary image distortion or inaccurate backlighting detection. Summary of the Invention

[0006] According to one embodiment of the method and apparatus for detecting backlight in an image, cumulative brightness data is generated based on multiple frames within a preset time period of the image, and a backlight score is calculated based on the cumulative brightness data to detect backlight in the image caused by sunlight from multiple light sources.

[0007] According to one embodiment of the method and apparatus for detecting backlight in an image, cumulative brightness data is generated based on multiple frames within a preset time period of the image, and the severity of the backlight in the image can be determined based on the cumulative brightness data.

[0008] The problems to be solved by this disclosure are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

[0009] According to one embodiment of the present invention, an apparatus for detecting backlight in an image is provided, comprising: a memory configured to store one or more instructions; and a processor configured to execute one or more instructions stored in the memory, wherein the processor converts pixel values ​​of each frame of an image in a plurality of frames within a preset time period based on a first threshold, divides the pixel values ​​into a plurality of block regions of the same size to generate a conversion frame, generates cumulative brightness data based on the conversion frame, determines the block region with the largest block value in the cumulative brightness data as a candidate backlight block region, classifies the plurality of block regions into a first region and a second region based on the position of the candidate backlight block regions, calculates a backlight score of the image using the sum of the block values ​​of the block regions of the first region and the sum of the block values ​​of the block regions of the second region, and determines the image as a backlight image when the backlight score is equal to or greater than a preset second threshold.

[0010] According to another embodiment of the present invention, a method for detecting backlight in an image is provided, comprising: converting pixel values ​​of each frame of an image within a preset time period based on a first threshold, and dividing the pixel values ​​into multiple blocks of the same size to generate a conversion frame; generating cumulative brightness data based on the conversion frame; determining the block region with the largest block value in the cumulative brightness data as a candidate backlight block region; classifying the multiple block regions into a first region and a second region based on the position of the candidate backlight block regions, calculating a backlight score of the image using the sum of the block values ​​of the block regions in the first region and the sum of the block values ​​of the block regions in the second region, and determining the image as a backlight image when the backlight score is equal to or greater than a preset second threshold.

[0011] According to one embodiment of the method and apparatus, the brightness accumulation data is generated based on multiple frames of an image over a predetermined time period, and the backlight score is calculated based on detecting backlight images generated by sunlight in various light sources using the brightness accumulation data, thereby improving the accuracy of backlight detection.

[0012] According to one embodiment of the method and apparatus, the severity of a backlight image is determined based on cumulative brightness data. Therefore, backlight situations where image recognition is impossible can be determined based on the severity of the backlight, thereby improving the reliability of the image recognition system. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of an apparatus for detecting backlighting in an image according to one embodiment of the present disclosure.

[0014] Figure 2A and 2B This is a schematic diagram illustrating the process of generating a conversion frame from an apparatus for detecting backlighting in an image according to one embodiment of the present disclosure.

[0015] Figure 3This is a schematic diagram illustrating the process of calculating the backlight score of an image by an apparatus for detecting backlight in an image according to one embodiment of the present disclosure.

[0016] Figure 4 This is a flowchart illustrating a method for detecting backlighting in an image according to one embodiment of the present disclosure. Detailed Implementation

[0017] Hereinafter, some exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In the following description, the same reference numerals preferably denote the same elements, even though the elements are shown in different drawings. Furthermore, in the following description of some embodiments, detailed descriptions of the known functions and configurations included therein will be omitted for the purpose of clarity and brevity.

[0018] Furthermore, terms such as first, second, A, B, (A), (B), etc., are used only to distinguish one component from another and do not imply or suggest the substance, order, or sequence of the components. In this specification, when a component “comprises” or “includes” another component, that component means to further include (but not exclude) other components, unless specifically stated to the contrary. Terms such as “unit” or “module” refer to one or more units for performing at least one function or operation, which can be implemented by hardware, software, or a combination thereof.

[0019] The following description of the present disclosure, taken in conjunction with the accompanying drawings, is intended to describe exemplary embodiments of the present disclosure and is not intended to represent the only embodiments in which the present disclosure may be practiced.

[0020] Figure 1 This is a schematic diagram of an apparatus for detecting backlighting in an image according to one embodiment of the present disclosure.

[0021] Reference Figure 1 An apparatus 100 for detecting backlight in an image according to one embodiment includes an input / output interface 110, a processor 120, and a memory 130. Here, the input / output interface 110, processor 120, and memory 130 of the apparatus 100 for detecting backlight in an image according to one embodiment can transmit data to each other via a bus 140.

[0022] According to one embodiment, the device 100 for detecting backlight in an image can be installed as a vehicle driving assistance device on a non-autonomous driving vehicle, or it can be installed on an autonomous driving vehicle for vehicle control.

[0023] The input / output interface 110 acquires data related to images captured by the vehicle's camera and provides this data to the processor 120. The data provided to the processor 120 by the input / output interface 110 may be multiple consecutive frames generated by the camera within each preset time period. Here, the multiple consecutive frames may be frames within the preset time period among multiple frames constituting an image.

[0024] The frames within a preset time period of the image can be frames generated during the time interval from a preset first time point to a second time point, but are not limited to this, and can be determined based on the number of frames. For example, based on a preset number of frames according to the camera's frame rate, the input / output interface 110 can provide a corresponding number of consecutive frames to the processor 120.

[0025] Processor 120 generates conversion frames based on multiple frames within a preset time period of the image. Specifically, processor 120 generates a corresponding conversion frame for each frame. Therefore, processor 120 generates the same number of conversion frames as the number of frames within the preset time period.

[0026] The processor 120 converts the pixel values ​​of the frame based on a first threshold. Here, the pixel values ​​can be pixel values ​​representing information about brightness from the pixel values ​​of multiple channels constituting the corresponding frame.

[0027] For example, when the frame is an RGB image, the pixel value can be the average of the pixel intensity in the R channel, the pixel intensity in the G channel, and the pixel intensity in the B channel. When the frame is a Hue / Saturation Value (HSV) image, the pixel value can be the pixel value of the V channel used for color values, and when the frame is a YUV image, the pixel value can be the pixel value of the Y channel used for illumination.

[0028] The first threshold is a preset value relative to the pixel value. The first threshold is set to a relatively high value within the range of pixel values ​​a pixel can have, in order to detect very bright light sources that can produce backlighting, such as sunlight. For example, in the case of an RGB image, the first threshold could be set to 220 within the pixel value range of 0 to 255.

[0029] The processor 120 converts the pixel values ​​of pixels in the frame that are less than a first threshold to 0. Therefore, pixels with pixel values ​​equal to or greater than the first threshold are extracted.

[0030] After the processor 120 converts the pixel values ​​of the frame, it divides the frame into multiple block regions of the same size to generate a converted frame. Here, the size of the block region is a preset size; for example, a block region is set to have a length corresponding to two pixels in the horizontal direction and two pixels in the vertical direction, respectively. Therefore, a block region can be set for a total of four pixels.

[0031] The size of the block area can be set based on the size of the sun as displayed on the frame. Since the sun is an object that is very far away from the camera, its size is captured almost constantly in the image captured by the camera. Specifically, the size of the sun being photographed increases with the lens focal length, but remains almost constant with changes in other shooting conditions. Therefore, the size of the block area can be set based on, but is not limited to, the size of the area corresponding to the sun appearing on the frame.

[0032] Each block region has a block value corresponding to it. Here, the block value is the average pixel value of multiple pixels corresponding to the block region. The processor 120 calculates the block value based on the pixel values ​​of the multiple pixels corresponding to the block region.

[0033] Processor 120 generates a cumulative luminance data based on multiple conversion frames. Here, the cumulative luminance data is a two-dimensional array data, which is generated based on values ​​obtained by accumulating the block values ​​of the luminance of each block region in multiple conversion frames over a preset time period for each block region.

[0034] The cumulative luminance data includes the same number and arrangement of block regions as the conversion frames. The processor 120 generates block values ​​for each block region of the cumulative luminance data based on multiple conversion frames over a preset time period.

[0035] Processor 120 sums the block values ​​of the block regions at the same location corresponding to each of the multiple conversion frames for each block region. Based on the sum of the values ​​for each block region, processor 120 generates a block value for the corresponding block region at the location in the cumulative luminance data.

[0036] For example, when the first conversion frame, the second conversion frame, and the third conversion frame are generated sequentially within a preset time period, the processor 120 sums the block value of the block region corresponding to the first position of the first conversion frame, the block value of the block region corresponding to the first position of the second conversion frame, and the block value of the block region corresponding to the first position of the third conversion frame, and generates the block value of the block region corresponding to the first position of the brightness accumulation data based on the sum.

[0037] The block value for a block region of luminance cumulative data can be, but is not limited to, the sum of block values ​​for corresponding positions in multiple transition frames, and can be the average of the block values. For example, processor 120 can generate a value obtained by dividing the sum of the block values ​​by the number of transition frames as the block value for the block region of luminance cumulative data.

[0038] When the position of the light source changes within a preset time period, it is displayed as a large block value in a block area at a different location in each transition frame. Therefore, it is difficult to accumulate block values ​​in the same block area when the position of the light source changes over time.

[0039] Meanwhile, if the position of the light source remains unchanged within a predetermined time period, for each transition frame, large block values ​​are repeatedly observed in block regions at the same or adjacent positions. Therefore, the block values ​​for light sources with unchanged positions are accumulated in the block regions at the same positions.

[0040] The result is that, in the cumulative luminance data, the block regions with large block values ​​are associated with light sources whose positions remain unchanged, and when the block values ​​are accumulated, the block values ​​for light sources whose positions change are relatively small compared to those for light sources whose positions remain unchanged.

[0041] Backlighting that interferes with vehicle image recognition during the day is mostly caused by sunlight. Generally speaking, in images captured by cameras of moving vehicles, sunlight is the only constant light source. Therefore, when a stationary light source is detected in the image, sunlight can be detected from the various light sources present on the road.

[0042] The processor 120 determines the block region with the maximum block value in the cumulative brightness data as a backlight candidate block region, and classifies multiple block regions of the cumulative brightness data into a first region and a second region based on the position of the backlight candidate block region.

[0043] The processor 120 identifies the block region with the largest block value among multiple block regions in the accumulated brightness data as a candidate backlight block region, and sets the candidate backlight block region and multiple adjacent block regions as a first region. For example, a total of nine block regions belonging to the range of three block regions in the horizontal direction and three block regions in the vertical direction centered on the candidate backlight block region can be set as the first region, but this disclosure is not limited thereto.

[0044] The processor 120 sets the remaining block regions outside the first region in the multiple block regions of the cumulative luminance data as the second region.

[0045] The processor 120 calculates the backlight score of the image by using the sum of the block values ​​of the block regions of the first region and the sum of the block values ​​of the block regions of the second region. Specifically, the backlight score of the image is calculated based on Equation 1.

[0046] Equation 1

[0047]

[0048] Here, score is the backlight score of the image, Light1 is the sum of the block values ​​of multiple block regions included in the first region, and Light2 is the sum of the block values ​​of multiple block regions included in the second region.

[0049] As the brightness difference between the first and second regions increases, the backlight score approaches 1. This increases the probability of backlighting. Therefore, when the backlight score approaches 1, the likelihood of a backlit image increases.

[0050] When the backlight score of an image is equal to or greater than a preset second threshold, the processor 120 determines that the image is a backlit image. Here, the second threshold is a value set based on the premise that the brightness difference between the first and second regions is large enough to cause backlighting in the image. The second threshold can be set to a value between 0 and 1. For example, the second threshold can be set to 0.7, but is not limited to this.

[0051] When an image is determined to be a backlit image, the processor 120 determines the severity of the backlighting based on the sum of block values ​​in a first region. Here, the severity of the backlighting can be determined as any one of two or more preset severity levels, depending on its impact on the vehicle's image recognition capabilities.

[0052] The processor 120 calculates the intensity of the backlight by using the sum of the block values ​​of the block regions in the first region. Specifically, the intensity of the backlight is calculated based on Equation 2.

[0053] Equation 2

[0054]

[0055] Here, degree is the intensity of the backlight. Light1 is the sum of the block values ​​of the first region's blocks, and Lightmax is the maximum value of the sum of the block values ​​of the first region's blocks. The maximum value of the sum of the block values ​​of the first region's blocks is calculated based on Equation 3.

[0056] Equation 3

[0057] Lightmax = N × Blockmax. Here, N is the number of block regions corresponding to the first region in the block region of the luminance accumulation data. Blockmax is the maximum block value that the block region of the luminance accumulation data can have. For example, when multiple conversion frames are generated based on an RGB image and the block value of the luminance accumulation data is the average of the block values ​​of the multiple conversion frames, the maximum block value that the block region of the luminance accumulation data can have is 255.

[0058] Processor 120 determines which of two or more preset backlight severity levels the intensity of the backlight belongs to.

[0059] For example, when the backlight intensity is 0.75 or higher, the preset backlight severity level can be set to high; when the backlight intensity is 0.3 or higher but less than 0.75, the preset backlight severity level can be set to medium; and when the backlight intensity is less than 0.3, the preset backlight severity level can be set to low. The processor 120 can determine the severity of the backlight as any one of high, medium, or low based on the backlight intensity.

[0060] The memory 130 may include volatile memory, permanent memory, virtual memory, or other types of memory for storing information used by or output by the device 100 for detecting backlight in an image. For example, the memory 130 may include random access memory (RAM) or dynamic RAM (DRAM).

[0061] The memory 130 may store programs for processing or controlling the processor 120 and various data for the operation of the device 100 for detecting backlight in an image. For example, the memory 130 may include at least one or more of the following: multiple frames of an image captured by the camera over a preset time period; multiple conversion frames and cumulative brightness data generated by the processor 120; and information regarding the severity of backlight in the image. Furthermore, the memory 130 may store setting information such as a first threshold, a second threshold, or two or more preset backlight severity levels.

[0062] Figure 2A and 2B This is a schematic diagram illustrating the process of generating a conversion frame using an apparatus for detecting backlighting in an image according to one embodiment of the present disclosure.

[0063] Figure 2A This is a schematic diagram of frame 200, in which the device for detecting backlight in an image converts pixel values ​​based on a first threshold. Figure 2B It is shown that by... Figure 2A The transformed frame 210 is a diagram generated by dividing frame 200 into multiple block regions.

[0064] Reference Figure 2A In frame 200, the pixel values ​​of all pixels with values ​​less than 220 (220 is the first threshold) are converted to 0. Therefore, pixels with pixel values ​​greater than 220 are extracted.

[0065] Pixels corresponding to areas of high brightness on the image frame (such as areas where light sources such as the sun and headlights of other vehicles are captured) are extracted, and the pixel values ​​corresponding to pixels in other areas are set to 0. Therefore, the sun pixel 201 and the headlight pixel 203 of other vehicles are extracted.

[0066] Simultaneously refer to Figure 2A and 2B The apparatus for detecting backlighting in an image divides the pixels of frame 200 into blocks of four pixels each, with each block consisting of four pixels. The conversion frame 210 includes a block region for each position corresponding to two pixels in the horizontal and vertical directions, thus totaling four pixels. Here, the block value of a block region is the average of the pixel values ​​of the four pixels corresponding to that block region.

[0067] The sun pixel 201 of frame 200 and the headlight pixel 203 of other vehicles are divided into sun block area 211 and headlight block area 213 of another vehicle at the corresponding positions in transition frame 210.

[0068] The device for detecting backlighting in an image generates a conversion frame 210 for each frame 200. Since the pixel values ​​of the pixels in each of the multiple frames constituting the image may be different, each of the multiple conversion frames can have a different block value arrangement.

[0069] Figure 3 This is a schematic diagram illustrating the process of calculating the backlight score of an image using an apparatus for detecting backlight in an image according to one embodiment of the present disclosure.

[0070] Reference Figure 3 The apparatus for detecting backlighting in an image generates cumulative brightness data 300 based on multiple conversion frames. Here, the cumulative brightness data 300 includes the number and arrangement of the same block regions as the conversion frames.

[0071] The block value of each block region in the cumulative luminance data 300 is the average of the block values ​​of the block regions at the same position corresponding to each transition frame. Here, the block regions with large block values ​​are for light sources with fixed positions, while the block regions with relatively small block values ​​are for light sources with changing positions.

[0072] The apparatus for detecting backlighting in an image identifies the block region with the maximum block value as a backlight candidate block region 311, and sets the backlight candidate block region 311 and eight block regions adjacent to it as a first region. The apparatus for detecting backlighting in an image sets the remaining block regions other than the first region 310 as a second region 320.

[0073] The device for detecting backlight in an image uses the sum of nine block values ​​in a first region and the sum of 31 block values ​​in a second region to calculate the backlight score of the image.

[0074] Figure 4 This is a flowchart illustrating a method for detecting backlighting in an image according to one embodiment of the present disclosure.

[0075] Reference Figure 4 The device for detecting backlighting in an image generates conversion frames for multiple frames within a preset time period of the image (S400). Here, the device for detecting backlighting in an image generates a corresponding conversion frame for each frame.

[0076] The apparatus for detecting backlighting in an image converts the pixel values ​​of a frame based on a preset first threshold. Here, the first threshold is a preset value for pixel values. Specifically, the first threshold is set to a relatively high value within the range of pixel values ​​a pixel can have, in order to detect strong light sources capable of generating backlighting. The apparatus for detecting backlighting in an image converts the pixel values ​​of pixels with values ​​less than the first threshold to 0.

[0077] An apparatus for detecting backlighting in an image generates a transformed frame by transforming the pixel values ​​of the frame and then dividing the frame into multiple block regions of equal size. Each block region has a corresponding block value. Here, the block value is the average of the pixel values ​​of the multiple pixels corresponding to the block region.

[0078] The apparatus for detecting backlighting in an image generates cumulative brightness data based on multiple conversion frames (S410). Here, the cumulative brightness data is a two-dimensional array of data, generated based on values ​​obtained by accumulating the block values ​​of the brightness of each block region in multiple conversion frames over a preset time period for each block region. The cumulative brightness data includes the same number and arrangement of block regions as the conversion frames.

[0079] The device for detecting backlight in an image sums the block values ​​of the block regions corresponding to the same position in each of the multiple conversion frames for each block region, and generates the block value of the corresponding position of the block region in the cumulative luminance data based on the sum of the block values ​​of each block region.

[0080] The block value of the luminance accumulation data block region can be, but is not limited to, the sum of the block values ​​of the corresponding block regions in multiple conversion frames, and can also be the average value of the block values.

[0081] The apparatus for detecting backlight in an image determines backlight candidate block regions from accumulated brightness data (S420). Here, the backlight candidate block regions are determined as the block regions with the largest block value among multiple block regions of accumulated brightness data.

[0082] The device for detecting backlight in an image classifies multiple block regions of accumulated brightness data into a first region and a second region, and calculates a backlight score (S430).

[0083] The device for detecting backlight in an image sets a candidate backlight block region and a plurality of adjacent block regions from a plurality of block regions of accumulated brightness data as a first region, and sets the remaining block regions outside the first region as a second region.

[0084] The apparatus for detecting backlighting in an image calculates a backlight score using the sum of block values ​​from a first region and the sum of block values ​​from a second region. The backlight score increases as the brightness difference between the first and second regions increases, and a higher backlight score indicates a more likely backlit image.

[0085] The apparatus for detecting backlight in an image determines whether the backlight score is equal to or greater than a preset threshold (S440). Here, the threshold may be a second threshold set based on the condition that the brightness difference between the first region and the second region is large enough to produce backlight in the image.

[0086] When the backlight score is equal to or greater than the second threshold, the device for detecting backlight in the image determines the image as a backlight image and determines the severity of the backlight (S450).

[0087] The device for detecting backlight in an image calculates the intensity of the backlight based on the sum of block values ​​of block regions in a first region, and determines whether the intensity of the backlight belongs to any one of two or more preset severity levels of backlight.

[0088] Each element of the apparatus or method according to the invention can be implemented in hardware or software, or a combination of hardware and software. The function of each element can be implemented in software, and a microprocessor can be implemented to execute the software functions corresponding to each element.

[0089] Various implementations of the systems and techniques described herein can be implemented using digital electronic circuits, integrated circuits, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. Various embodiments may include implementations having one or more computer programs executable on a programmable system. The programmable system includes at least one programmable processor, which may be a dedicated or general-purpose processor, coupled to receive and send data and instructions to a storage system, at least one input device, and at least one output device. The computer program (also referred to as a program, software, software application, or code) includes instructions for the programmable processor and is stored in a computer-readable recording medium.

[0090] Computer-readable recording media can include all types of storage devices on which computer-readable data can be stored. Computer-readable recording media can be non-volatile or non-transitory media, such as read-only memory (ROM), random access memory (RAM), optical disc ROM (CD-ROM), magnetic tape, floppy disk, or optical data storage devices. Additionally, computer-readable recording media can also include transitional media such as data transmission media. Furthermore, computer-readable recording media can be distributed across computer systems connected via a network, and computer-readable program code can be stored and executed in a distributed manner.

[0091] Although the flowcharts / timing diagrams in this specification show operations executed sequentially, this is merely an exemplary description of the technical concept of one embodiment of this disclosure. In other words, those skilled in the art to which one embodiment of this disclosure pertains will understand that various modifications and changes can be made without departing from the essential characteristics of the embodiments of this disclosure; that is, the order shown in the flowcharts / timing diagrams can be changed, and one or more operations can be executed in parallel. Therefore, the flowcharts / timing diagrams are not limited to a temporal order.

[0092] Although exemplary embodiments of this disclosure have been described for illustrative purposes, those skilled in the art will understand that various modifications, additions, and substitutions are possible without departing from the spirit and scope of the claimed invention. Therefore, exemplary embodiments of this disclosure have been described for the sake of brevity and clarity. The scope of the technical concept of these embodiments is not limited by the illustrations. Therefore, those skilled in the art will understand that the scope of the claimed invention is not limited to the embodiments explicitly described above, but is limited by the claims and their equivalents.

Claims

1. An apparatus for detecting backlighting in an image, comprising: The memory is configured to store one or more instructions; as well as The processor is configured to execute one or more of the instructions stored in the memory. Wherein, the processor: Obtain the pixel values ​​of the pixels in each of a series of image frames generated consecutively within a preset time period. If the pixel value is less than the first threshold, then the pixel value is converted to 0. The frame is divided into multiple block regions of the same size to generate a converted frame, wherein the size of the block region is determined based on the size of the sun displayed on the frame. Luminance accumulation data is generated based on the converted frame. The block region with the largest block value in the accumulated brightness data is determined as the candidate backlight block region. Based on the location of the backlight candidate block regions, the plurality of block regions are classified into a first region and a second region. The backlight score of the image is calculated using the sum of the block values ​​of the block regions in the first region and the sum of the block values ​​of the block regions in the second region. Based on the backlight score and the second threshold, the image is determined to be a backlight image. The first threshold is set as a value used to detect sunlight.

2. The apparatus according to claim 1, wherein, When the image is determined to be a backlit image, the processor determines the severity of the backlighting based on the sum of the block values ​​of the block regions in the first region.

3. The apparatus according to claim 1, wherein, The pixel value is a brightness-related pixel value among the pixel values ​​associated with multiple channels of the image frame.

4. The apparatus according to claim 1, wherein, The processor divides the image frame into multiple block regions of a preset size, and calculates the block value for each block region. The block value is the average pixel value of the pixels corresponding to the block region.

5. The apparatus according to claim 1, wherein, The number of block regions in the accumulated luminance data is the same as the number of block regions in the converted frame, and The block value of the block region of the luminance accumulation data is a value generated by adding the block values ​​of the block regions at the same positions corresponding to the positions of the block regions of the luminance accumulation data in the conversion frame for each block region.

6. The apparatus according to claim 1, wherein, The first region includes the backlight candidate block region and multiple block regions located adjacent to the backlight candidate block region among the multiple block regions included in the brightness accumulation data, and The second region includes the remaining block regions included in the accumulated brightness data.

7. The apparatus according to claim 1, wherein, The backlight score of the image is calculated based on the relative ratio between the sum of the block values ​​of the block regions in the first region and the sum of the block values ​​of the block regions in the second region.

8. The apparatus according to claim 2, wherein, The processor calculates the intensity of the backlight based on the sum of the block values ​​of the block regions in the first region, and determines which of two or more preset severity levels the image belongs to based on the intensity of the backlight.

9. A method for detecting backlighting in an image, comprising: Obtain the pixel values ​​of the pixels in each of a series of image frames generated consecutively within a preset time period; If the pixel value is less than the first threshold, then the pixel value is converted to 0. The frame is divided into multiple block regions of the same size to generate a converted frame, wherein the size of the block region is determined based on the size of the sun displayed on the frame. Luminance accumulation data is generated based on the converted frame. The block region with the largest block value in the accumulated brightness data is determined as the candidate backlight block region. Based on the location of the backlight candidate block regions, the plurality of block regions are classified into a first region and a second region. The backlight score of the image is calculated using the sum of the block values ​​of the block regions in the first region and the sum of the block values ​​of the block regions in the second region. Based on the backlight score and the second threshold, the image is determined to be a backlight image. The first threshold is set as a value used to detect sunlight.

10. The method of claim 9, further comprising: When the image is determined to be a backlit image, the severity of the backlight is determined based on the sum of the block values ​​of the block regions in the first region.

11. The method according to claim 9, wherein, The pixel value is a brightness-related pixel value among the pixel values ​​associated with multiple channels of the image frame.

12. The method according to claim 9, wherein, Generating the converted frame includes dividing the image frame into multiple block regions of a preset size, and calculating the block value for each block region. The block value is the average pixel value of the pixels corresponding to the block region.

13. The method according to claim 9, wherein, The number of block regions in the accumulated luminance data is the same as the number of block regions in the converted frame, and The block value of the block region of the luminance accumulation data is a value generated by adding the block values ​​of the block regions at the same positions corresponding to the positions of the block regions of the luminance accumulation data in the conversion frame for each block region.

14. The method according to claim 9, wherein, The first region includes the backlight candidate block region and multiple block regions located adjacent to the backlight candidate block region among the multiple block regions included in the brightness accumulation data, and The second region includes the remaining block regions included in the accumulated brightness data.

15. The method according to claim 9, wherein, The backlight score of the image is calculated based on the relative ratio between the sum of the block values ​​of the block regions in the first region and the sum of the block values ​​of the block regions in the second region.

16. The method of claim 10, wherein, The intensity of the backlight is calculated based on the sum of the block values ​​of the block regions in the first region, and the image is determined to belong to one of two or more preset severity levels based on the intensity of the backlight.

Images