WO2021138988A1 - 一种设定成像参数的方法、系统及计算机可读存储介质 - Google Patents

一种设定成像参数的方法、系统及计算机可读存储介质 Download PDF

Info

Publication number
WO2021138988A1
WO2021138988A1 PCT/CN2020/077859 CN2020077859W WO2021138988A1 WO 2021138988 A1 WO2021138988 A1 WO 2021138988A1 CN 2020077859 W CN2020077859 W CN 2020077859W WO 2021138988 A1 WO2021138988 A1 WO 2021138988A1
Authority
WO
WIPO (PCT)
Prior art keywords
imaging
interest
imaging parameters
region
parameters
Prior art date
Application number
PCT/CN2020/077859
Other languages
English (en)
French (fr)
Inventor
张敏
王卫芳
朱毅博
王闯闯
胡正
Original Assignee
深圳奥比中光科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳奥比中光科技有限公司 filed Critical 深圳奥比中光科技有限公司
Publication of WO2021138988A1 publication Critical patent/WO2021138988A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • H04N23/64Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70Circuitry for compensating brightness variation in the scene
    • H04N23/73Circuitry for compensating brightness variation in the scene by influencing the exposure time

Definitions

  • the present invention relates to the technical field of setting imaging parameters, in particular to a method, system and computer-readable storage medium for setting imaging parameters.
  • the quality of the imaging environment is critical to the inspection results. Whether the imaging parameters are appropriate or not directly affects the texture, brightness, clarity, and color of the image. For example, unreasonable imaging parameter settings can lead to overexposure or underexposure, both of which can lead to a lot of loss of detailed information. Therefore, in an imaging system, in order to construct a high-quality imaging environment, it is often necessary to select appropriate imaging parameters.
  • Industrial cameras generally use automatic or manual adjustments for imaging parameters.
  • industrial cameras that automatically adjust imaging parameters will quantify the gray value, judge the intensity of the light to a certain extent, and base on the preset threshold. Judge overexposure or underexposure of light, and then adjust the exposure time, gain and other parameters to complete the photo.
  • This method can make the overall image overexposed and not underexposed, but for multiple local areas of interest in the image, it is impossible to achieve one.
  • Manual adjustment can shoot better quality images, but this method requires manual adjustment of multiple image parameter combinations, which is time-consuming and labor-intensive. For non-professionals, there is a certain degree of adjustment in parameters, selection of parameter combinations, and judgment of high-quality images. Difficulties.
  • the present invention provides a method, a system and a computer-readable storage medium for setting imaging parameters.
  • a method for setting imaging parameters includes the following steps: S1: traversing and acquiring images containing multiple regions of interest under different imaging parameters of the camera and extracting the number of feature points of the regions of interest in each image; S2: Sort each of the imaging parameters according to the number of feature points of the region of interest, and obtain a sorting result of each region of interest under each imaging parameter; S3: according to the sorting result and advance The imaging parameter combination is filtered out by the set number of imaging parameters; S4: determining the detected region of interest corresponding to each imaging parameter in the imaging parameter combination.
  • a voting method is used to screen out imaging parameter combinations based on the sorting result and the preset number of imaging parameters.
  • the screening of imaging parameter combinations includes the following steps: according to the ranking result, the imaging parameters of each region of interest in each ranking are calculated, and statistics are calculated for each ranking.
  • the frequency of each imaging parameter is sorted from high to low in a way that the importance of each imaging parameter gradually decreases from the first to the last to obtain the frequency ranking; if there is no even number, follow the frequency
  • the imaging parameters of the preset number of imaging parameters are selected from high to low to form the imaging parameter combination; if there is a flat number, the second to the last of the flat number appears in the frequency ordering
  • the reordering is carried out in a manner in which the importance gradually decreases, until the imaging parameters of the preset number of imaging parameters with a high frequency are selected to form the imaging parameter combination.
  • screening out imaging parameter combinations includes the following steps: obtaining the top k groups of imaging parameters for each region of interest according to the sorting result, where k ⁇ 2; The imaging parameters of each of the regions of interest under each ranking, and the frequency of each imaging parameter under each ranking, and the frequency is calculated in a way that the importance gradually decreases from the first to the last.
  • determining the region of interest detected corresponding to each of the imaging parameters in the imaging parameter combination includes: if the region of interest is ranked first in the ranking result of the region of interest If the imaging parameter is in the imaging parameter combination, the first-ranked imaging parameter is selected to detect the region of interest; if the first-ranked imaging parameter of the region of interest is not in the imaging parameter combination, then the imaging parameter is selected according to The imaging parameter with the higher order in the ranking result detects the region of interest.
  • the range of the imaging parameter is determined according to the initial value, the end value and the step size of the exposure and gain.
  • An algorithm is used to frame the region of interest.
  • a feature detection algorithm including scale-invariant feature transformation and accelerated robustness features is used to obtain the number of feature points in the region of interest.
  • the present invention also provides a system for setting imaging parameters, including: a camera for acquiring images containing multiple regions of interest under different imaging parameters, and transmitting the images to a processor; the processor uses To perform the following functions: receiving the image and extracting the number of feature points of each region of interest; sorting each of the imaging parameters according to the number of feature points of the region of interest to obtain each of the feelings The sorting result of the region of interest under each imaging parameter; the imaging parameter combination is screened out according to the sorting result and the preset number of imaging parameters; it is determined that each imaging parameter in the imaging parameter combination corresponds to the detected interest area.
  • the present invention further provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above methods are realized.
  • the beneficial effect of the present invention is to provide a method, system and computer-readable storage medium for setting imaging parameters.
  • marking the region of interest only the features of the region of interest in images with different imaging parameters are extracted, and based on the features Select the optimal imaging parameter for each region of interest, and use a voting mechanism to obtain an imaging parameter combination that includes at least one set of optimal imaging parameters.
  • the imaging parameter combination that is suitable for detection under different imaging parameters is given.
  • the region of interest can effectively solve the problem of selecting the best exposure time and gain parameters after the camera's aperture and focal length are adjusted to reasonable values. Taking into account the differences in light adaptability of different regions of interest, one or more can be selected.
  • the imaging parameter combination composed of three imaging parameters, and the region of interest recommended for detection under different imaging parameters can effectively adapt to different detection requirements.
  • the mechanism of giving imaging parameters based on the number of feature points in the region of interest is in line with the human visual attention mechanism, that is, more attention is focused on the characteristics of the region of interest, so the combination of imaging parameters set based on the number of feature points is also More reasonable.
  • the imaging quality is evaluated by an unsupervised method, and the evaluation criteria are unified. Under the same input conditions, the uniqueness of the results can be guaranteed, and the inconsistency of the parameter setting results caused by the difference of human subjective evaluation standards can be avoided. Uniqueness avoids the impact on data collection due to different evaluations of imaging quality by random personnel.
  • the operation is simple and efficient, and it can be operated by non-professionals. Users only need to set the start and end values of the exposure time, gain and other parameters, as well as the step length, and then they can shoot multiple images under multiple imaging, and then give them suitable for all A set of optimal imaging parameters for the region of interest, and multiple imaging parameters can also be given to suit different regions of interest.
  • Fig. 1 is a schematic diagram of a system for setting imaging parameters in an embodiment of the present invention.
  • Fig. 1 is a schematic diagram of a method for setting imaging parameters in an embodiment of the present invention.
  • Fig. 3 is a schematic flowchart of a method for setting imaging parameters in an embodiment of the present invention.
  • connection can be used for fixing or circuit connection.
  • first and second are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with “first” and “second” may explicitly or implicitly include one or more of these features.
  • a plurality of means two or more than two, unless otherwise specifically defined.
  • FIG. 1 is a schematic structural diagram of a system 10 for setting imaging parameters in an embodiment of the present invention.
  • the imaging parameter system 10 includes a camera 11 and a processor 12.
  • the camera 11 is used to take multiple images containing multiple regions of interest under different imaging parameters, and transmit the images to the processor 12;
  • the processor 12 receives the images and extracts the number of feature points of each region of interest in the image, according to The number of feature points of the region of interest is sorted for each imaging parameter to obtain the sorting result of each region of interest under each imaging parameter; the imaging parameters are filtered out according to the sorting result and the preset number of imaging parameters Combination; determining the detected region of interest corresponding to each imaging parameter in the imaging parameter combination.
  • the camera 11 takes an image and transmits the image to the processor 12.
  • the processor 12 receives the image and extracts the region of interest, according to the initial value, end value, and step size of imaging parameters such as exposure time and gain. Determine the combination of imaging parameters. It is understandable that the smaller the initial value, the larger the end value, and the smaller the step size, the number of combinations of imaging parameters will be more, that is, the selected imaging parameters will be more reasonable.
  • any set of imaging parameters (including exposure and gain), shoot the object to be detected, mark m regions of interest in the image, and specify the range of imaging parameters to be searched, that is, the initial values of exposure and gain, etc. , End value and step length, and the number of imaging parameters to be set n. It is understandable that the values of m and n are integers greater than or equal to 1. Traverse the combination of exposure time and gain imaging parameters provided, and shoot multiple images under different imaging parameters to obtain the regions of interest corresponding to multiple images with different imaging parameters. Use feature detection algorithms to extract the features of each region of interest, calculate the number of feature points in each region of interest in different imaging parameter images, and sort the imaging parameters based on the number of feature points in each region of interest.
  • the region of interest can be manually framed or recommended by algorithm.
  • scale-invariant feature transform SIFT
  • accelerated robust features Speeded Up Robust Features, SURF
  • other feature detection algorithms there is no restriction on the method of obtaining the number of feature points.
  • the imaging parameter combination composed of n component imaging parameters can be selected through a voting method. According to the sorting results, the imaging parameters of each region of interest under each ranking are counted, and the frequency of each imaging parameter under each ranking is counted, and the importance of the frequency is gradually reduced from the first to the last. Sort from high to low to get the frequency ranking; if there is no flat number, select the preset imaging parameter number from high to low according to the frequency ranking to form the imaging parameter combination;
  • the imaging parameter with the preset number of imaging parameters with high frequency Compose the imaging parameter combination. For example, if there is a tie in the first place, according to the frequency of the imaging parameter in the second place, and the imaging parameter with the largest frequency in the second place is selected, if the frequency in the second place is compared, the tie will appear again, Continue to compare the frequency of flat imaging parameters in the third place until the ranking of flat imaging parameters can be distinguished, from which the best n imaging parameters are selected and the best imaging parameter set is formed.
  • Table 1 there are 9 component image parameters and 6 regions of interest.
  • the first region of interest has the largest number of feature points in the image obtained under the fourth component image parameter, that is, the fourth component image parameter is the first 1st place in an area of interest.
  • the fourth component image parameter is the first of the second region of interest
  • the 9th component image parameter is the first of the third region of interest
  • the fourth component image parameter is the fourth region of interest.
  • the 6th component image parameter is the first place of the 5th region of interest
  • the 9th component image parameter is the first place of the 6th region of interest.
  • the fourth component image parameter is the first of the three regions of interest
  • the ninth component image parameter is the first of the two regions of interest
  • the sixth component image parameter is the first of the three regions of interest.
  • the 4th and 9th group image parameters use the 4th group image parameters to take the first image, which is suitable for counting the imaging information of the first, second, and fourth regions of interest in the image
  • the ninth component image parameter takes the second image, which is suitable for statistics of the imaging information of the third and sixth regions of interest in the image, and the fifth region of interest has more feature points due to the fourth component image parameter.
  • the imaging parameters of the same ranking may have a tie, and it is necessary to compare the frequency of the imaging parameters of the tie in the next ranking until the ranking of the imaging parameters of the tie can be distinguished.
  • Table 2 There are 9 component image parameters and 6 regions of interest in Table 2.
  • the first region of interest has the largest number of feature points in the image obtained under the fourth component image parameter, that is, the fourth component image parameter is the first 1st place in an area of interest.
  • the fourth component image parameter is the first of the second region of interest
  • the ninth component image parameter is the first of the third region of interest
  • the first component image parameter is the fourth region of interest.
  • the 6th component image parameter is the first place of the 5th region of interest
  • the 9th component image parameter is the first place of the 6th region of interest.
  • the fourth component image parameter is the first of the two regions of interest
  • the ninth component image parameter is the first of the two regions of interest
  • the first component image parameter is the first of the two regions of interest.
  • the third component image parameter is the second of the first region of interest
  • the fifth component image parameter is the second of the second region of interest
  • the 8th component image parameter is the third of interest
  • the fifth component image parameter is the second place of the fourth area of interest
  • the seventh component image parameter is the second place of the fifth area of interest
  • the first component image parameter is the sixth sense. The second place in the area of interest. Since the 4th and 9th group image parameters did not appear, it is still in a tie state, so it is necessary to count the frequency of the 4th and 9th group image parameters appearing in the third place.
  • the 5th component image parameter is the third place of the first region of interest
  • the 6th component image parameter is the second impression.
  • the third place of the area of interest, the seventh component image parameter is the third place of the third area of interest, the seventh component image parameter is the third place of the fourth area of interest, and the fourth component image parameter is the fifth
  • the 8th group image parameter is the third place of the 6th region of interest. Because the fourth component image parameter is the third of the fifth region of interest, and the ninth component image parameter does not appear, the fourth component image parameter is the first.
  • Group imaging parameters use the fourth component imaging parameter to take the first image, which is suitable for counting the imaging information of the first, second, and fourth regions of interest in the image, and use the ninth component imaging parameter to take the second image , Suitable for statistics of the imaging information of the 3rd and 6th regions of interest in the image, and the 5th region of interest has more feature points in the 4th composition image parameter, that is, the first image is selected for statistics
  • the imaging information of the fifth region of interest can be calculated under suitable imaging parameters for different regions of interest, thereby improving the imaging quality of the regions of interest.
  • another voting method can be used to filter out the n composition image parameters.
  • the frequency of, and according to the way of decreasing importance from the first to the last, the frequency is sorted from high to low to get the frequency ranking; if there is no even number, select the frequency from high to low according to the frequency ranking
  • the imaging parameters of the preset number of imaging parameters constitute the imaging parameter combination; if there is a flat number, the order will be sequentially reordered according to the way in which the importance of the second to the last of the flat numbers in the frequency ranking gradually decreases , Until the imaging parameters of the preset number of imaging parameters with high frequency are selected to form the imaging parameter combination.
  • the first K groups of imaging parameters (K greater than or equal to 2) with the largest number of feature points in each region of interest are obtained according to the order of the number of feature points, that is Say, 6 regions of interest can vote on 9 image parameters, and each region of interest can vote 2 votes, that is, there are 12 votes in total.
  • the first region of interest has the largest number of feature points in the image obtained under the third and fourth group image parameters, that is, the first region of interest will vote for the third and fourth groups.
  • the second area of interest will vote for groups 4 and 5
  • the third area of interest will vote for groups 8 and 9
  • the fourth area of interest will vote for groups. Vote for groups 4 and 5.
  • the fifth area of interest will vote for groups 6 and 7, and the sixth area of interest will vote for groups 9 and 8.
  • the 4th group image parameter gets 3 votes
  • the 5th group, the 8th group and the 9th group image parameters each get 2 votes
  • the 3rd, the 7th group and the 8th group image parameters each get 1 vote.
  • the 5th, 8th, and 9th group image parameters of each region of interest vote for the 5th, 8th and 9th groups, then the first region of interest will be Vote for group 5, the second area of interest will vote for group 5, the third area of interest will vote for group 9, and the fourth area of interest will vote for group For groups 5, the 5th region of interest will vote for the 5th group, and the 6th region of interest will vote for the 9th group. Then the 5th group gets 4 votes and the 9th group gets 2 votes.
  • the second image is taken using the fifth component imaging parameter, which is suitable for statistics of the imaging information of the third and sixth regions of interest in the image
  • the fifth region of interest is due to the fourth component of the imaging parameter.
  • the number of feature points is larger, that is, the first image is selected to count the imaging information of the fifth region of interest. In this way, the imaging information can be calculated under suitable imaging parameters for different regions of interest, thereby improving the imaging quality of the regions of interest.
  • n is set according to the needs of the user. You can choose 1 group imaging parameter shooting, or you can choose multiple group imaging parameter shooting. Users can select suitable imaging parameters for shooting according to their own region of interest.
  • Fig. 2 is a flowchart of a method for setting imaging parameters in an embodiment of the present invention, which includes the following steps:
  • S1 Traverse to obtain images containing multiple regions of interest under different imaging parameters of the camera, and extract the number of feature points of the region of interest in each image;
  • the camera 11 takes multiple images containing multiple regions of interest under different imaging parameters, and transmits the images to the processor 12, and the processor 12 receives the images and uses a feature detection algorithm to extract the number of feature points of the region of interest in the image. Item.
  • S2 Sort each of the imaging parameters according to the number of feature points of the region of interest, to obtain a sorting result of each region of interest under each of the imaging parameters;
  • the processor 12 sorts the imaging parameters of each region of interest according to the number of feature points of the region of interest, and counts the frequency of appearance of each imaging parameter in each ranking.
  • the frequency of occurrence of the imaging parameters is sorted from high to low, and the imaging parameter combinations are filtered out.
  • the present invention implements all or part of the processes in the above-mentioned embodiment methods, and can also be completed by instructing relevant hardware through a computer program.
  • the computer program can be stored in a computer-readable storage medium, and the computer program is executed by the processor. When executed, the steps of the foregoing method embodiments can be realized.
  • the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate forms.
  • the computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media, etc.
  • ROM Read-Only Memory
  • RAM Random Access Memory
  • electrical carrier signal telecommunications signal
  • software distribution media etc.
  • the content contained in the computer-readable medium can be appropriately added or deleted according to the requirements of the legislation and patent practice in the jurisdiction.
  • the computer-readable medium Does not include electrical carrier signals and telecommunication signals.
  • the present invention evaluates the image imaging quality according to the number of feature points of the region of interest in the image, and provides optimal imaging parameters suitable for all regions of interest, and can also provide multiple imaging parameters to suit different regions of interest respectively. Similar to the human visual cognitive system, the better the general lighting environment, the more details of the target that the human eye can observe, and the more feature points extracted in the image, the better the lighting environment of the image, so according to the characteristics The image under the best imaging parameters recommended by the number of points is consistent with the high-quality image selected visually.
  • the user selects any set of imaging parameters (including exposure and gain) according to actual needs, photographs the object to be detected, selects the region of interest in the image, and Given the range of imaging parameters to be searched, that is, the initial value, end value and step length of exposure and gain, etc., as well as the number of imaging parameters to be set; when the camera's aperture and focal length are adjusted to appropriate values, the camera automatically shoots a series
  • the image under the combination of imaging parameters (exposure, gain) is calculated by calculating the number of feature points in the image area of interest, and the imaging parameters are sorted based on the number of feature points, and a reasonable combination of imaging parameters suitable for the area of interest is recommended by voting.
  • the user can further confirm the selection according to the recommended imaging parameter combination, and then take the desired image.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Studio Devices (AREA)

Abstract

本发明提供一种设定成像参数的方法、系统及计算机可读存储介质,方法,包括:遍历获取相机在不同成像参数下的包含多个感兴趣区域的图像并提取每个图像中所述感兴趣区域的特征点数目;根据所述感兴趣区域的特征点数目的多少对每个所述成像参数进行排序,得到每个所述感兴趣区域在每个所述成像参数下的排序结果;根据所述排序结果和预先设置的成像参数数目筛选出成像参数组合;确定所述成像参数组合中每一个成像参数对应检测的所述感兴趣区域。可以有效解决相机在光圈、焦距调节至合理值后,最佳曝光时间和增益参数等参数的选择问题,可有效适应不同的检测需求。

Description

一种设定成像参数的方法、系统及计算机可读存储介质 技术领域
本发明涉及设定成像参数技术领域,尤其涉及一种设定成像参数的方法、系统及计算机可读存储介质。
背景技术
在工业视觉检测中,成像环境的质量对检测结果至关重要。成像参数的恰当与否,直接影响图像的质感、亮度、清晰度、色彩等。比如,成像参数设置的不合理会导致曝光过度或者曝光不足,这两者均会导致大量的细节信息丢失。因此在成像系统中,为了构建高质量的成像环境,往往需要选取合适的成像参数。
工业相机对于成像参数的设定一般采用自动调整或手动调整,一般采用自动调整成像参数的工业相机会进行灰度值的量化,对光线的强弱进行一定程度的判断,并根据预设的阈值判断光线过曝或光线不足,进而通过调整曝光时间、增益等参数来完成拍照,该方法可以使图像整体上不过曝,不欠曝,但是对于图像中多个局部感兴趣区域,无法做到一一满足。手动调整能够拍摄质量较优的图像,但是该方法需要人为调节多组成像参数组合,耗时耗力,对于非专业人士来说,在调节参数、选取参数组合、判断优质图像方面,会存在一定的困难。
以上背景技术内容的公开仅用于辅助理解本发明的构思及技术方案,其并不必然属于本专利申请的现有技术,在没有明确的证据表明上述内容在本专利申请的申请日已经公开的情况下,上述背景技术不应当用于评价本申请的新颖性和创造性。
发明内容
本发明为了解决现有的问题,提供一种设定成像参数的方法、系统及计算机可读存储介质。
为了解决上述问题,本发明采用的技术方案如下所述:
一种设定成像参数的方法,包括如下步骤:S1:遍历获取相机在不同成像参数下的包含多个感兴趣区域的图像并提取每个图像中所述感兴趣区域的特征点数目;S2:根据所述感兴趣区域的特征点数目的多少对每个所述成像参数进行排序,得到每个所述感兴趣区域在每个所述成像参数下的排序结果;S3:根据所述排序结果和预先设置的成像参数数目筛选出成像参数组合;S4:确定所述成像参 数组合中每一个成像参数对应检测的所述感兴趣区域。
在本发明的一种实施例中,采用投票法根据所述排序结果和所述预先设置的成像参数数目筛选出成像参数组合。
在本发明的又一种实施例中,筛选出成像参数组合包括如下步骤:根据所述排序结果统计出每个名次下每个所述感兴趣区域的成像参数,并统计每个所述名次下每个所述成像参数的频数,并按照从第一名到最后一名重要性逐渐降低的方式,对频数进行从高到低地排序,得到频数的排序;若没有平数,则按照所述频数的排序从高到低的选取所述预先设置的成像参数数目的成像参数组成所述成像参数组合;若有平数,则根据所述频数的排序中出现平数的后面一名到最后一名重要性逐渐降低的方式依次进行重新排序,直至选出频数高的所述预先设置的成像参数数目的成像参数组成所述成像参数组合。
在本发明的再一种实施例中,筛选出成像参数组合包括如下步骤:根据所述排序结果获取到每个所述感兴趣区域前k组的成像参数,其中,k≥2;统计出每个名次下每个所述感兴趣区域的成像参数,并统计每个所述名次下每个所述成像参数的频数,并按照从第一名到最后一名重要性逐渐降低的方式,对频数进行从高到低地排序,得到频数的排序;若没有平数,则按照所述频数的排序从高到低的选取所述预先设置的成像参数数目的成像参数组成所述成像参数组合;若有平数,则根据所述频数的排序中出现平数的后面一名到最后一名重要性逐渐降低的方式依次进行重新排序,直至选出频数高的所述预先设置的成像参数数目的成像参数组成所述成像参数组合。
在本发明的又一种实施例中,确定所述成像参数组合中每一个所述成像参数对应检测的所述感兴趣区域包括:若所述感兴趣区域的所述排序结果中排名第一的成像参数在所述成像参数组合中,则选择所述排名第一的成像参数检测所述感兴趣区域;若所述感兴趣区域的排名第一的成像参数不在所述成像参数组合中,则按照所述排序结果中的顺序靠前的所述成像参数检测所述感兴趣区域。所述成像参数的范围根据曝光和增益的初始值、结束值以及步长确定。通过算法框出所述感兴趣区域。采用包括尺度不变性特征变换、加速鲁棒性特征的特征检测算法获取所述感兴趣区域的特征点数目。
本发明还提供一种设定成像参数系统,包括:相机,用于获取在不同成像参 数下的包含多个感兴趣区域的图像,并将所述图像传至处理器;所述处理器,用于执行以下功能:接收所述图像并提取每个所述感兴趣区域的特征点数目;根据所述感兴趣区域的特征点数目的多少对每个所述成像参数进行排序,得到每个所述感兴趣区域在每个所述成像参数下的排序结果;根据所述排序结果和预先设置的成像参数数目筛选出成像参数组合;确定所述成像参数组合中每一个成像参数对应检测的所述感兴趣区域。
本发明再提供一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序被处理器执行时实现如上任一所述方法的步骤。
本发明的有益效果为:提供一种设定成像参数的方法、系统及计算机可读存储介质,通过通过标示出感兴趣区域,仅提取不同成像参数的图像中感兴趣区域的特征,并根据特征数目的多少选出针对每个感兴趣区域的最佳成像参数,并采用投票机制得出包括至少一组最佳成像参数的成像参数组合,同时给定成像参数组合中不同成像参数下适宜检测的感兴趣区域,可以有效解决相机在光圈、焦距调节至合理值后,最佳曝光时间和增益参数等参数的选择问题,并考虑到不同感兴趣区域对光照适应性差异,可选出一个或多个成像参数组成的成像参数组合,以及在不同成像参数下建议检测的感兴趣区域,可有效适应不同的检测需求。
基于感兴趣区域的特征点数目给出成像参数的机制,契合人的视觉注意力机制,即将更多的注意力集中在感兴趣区域的特性,所以基于特征点数目排序设定的成像参数组合也更合理。
选取具有代表性的感兴趣区域的特征点而不是对整幅图像统计信息,更能反应局部的成像质量。
采用无监督的方法对成像质量进行评价,且评判标准统一,在相同的输入条件下可保证结果的唯一性,避免了由于人的主观评判标准的差异性而带来的参数设定结果的不唯一性,避免了因随机人员对成像质量评价的不同而对数据采集产生影响。
从根本上避免了人工手动调节成像参数,并在多种成像参数中选择适合不同感兴趣区域检测的繁琐过程,实现了不同感兴趣区域设定不同成像参数的功能。
操作简易且高效,非专业人士也可操作,用户仅需要设置曝光时间、增益等参数的起始与终止值,以及步长,即可拍摄多个成像下的多张图像,进而给出适 合所有感兴趣区域的一组最佳成像参数,也可以给出多组成像参数以分别适合不同感兴趣区域。
附图说明
图1是本发明实施例中设定成像参数的系统示意图。
图1是本发明实施例中设定成像参数的方法示意图。
图3是本发明实施例中设定成像参数的方法的流程示意图。
其中,10-成像参数系统,11-相机,12-处理器。
具体实施方式
为了使本发明实施例所要解决的技术问题、技术方案及有益效果更加清楚明白,以下结合附图及实施例,对本发明进行进一步详细说明。应当理解,此处所描述的具体实施例仅仅用以解释本发明,并不用于限定本发明。
需要说明的是,当元件被称为“固定于”或“设置于”另一个元件,它可以直接在另一个元件上或者间接在该另一个元件上。当一个元件被称为是“连接于”另一个元件,它可以是直接连接到另一个元件或间接连接至该另一个元件上。另外,连接既可以是用于固定作用也可以是用于电路连通作用。
需要理解的是,术语“长度”、“宽度”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系,仅是为了便于描述本发明实施例和简化描述,而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作,因此不能理解为对本发明的限制。
此外,术语“第一”、“第二”仅用于描述目的,而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此,限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多该特征。在本发明实施例的描述中,“多个”的含义是两个或两个以上,除非另有明确具体的限定。
图1为本发明一个实施例中一种设定成像参数系统10的结构示意图。成像参数系统10包括相机11、处理器12。相机11用于拍摄多张不同成像参数下包含多个感兴趣区域的图像,并将图像传至处理器12;处理器12接收该图像并提取图像的每个感兴趣区域的特征点数目,根据所述感兴趣区域的特征点数目的多少对每个所述成像参数进行排序,得到每个感兴趣区域在每个成像参数下的排序 结果;根据排序结果和预先设置的成像参数数目筛选出成像参数组合;确定成像参数组合中每一个成像参数对应检测的所述感兴趣区域。
在一个实施例中,相机11拍摄图像,并将图像传至处理器12,处理器12接收该图像并提取出感兴趣区域,根据曝光时间和增益等成像参数的初始值、结束值以及步长确定成像参数的组合。可以理解的是,初始值越小,结束值越大,步长越小,那么成像参数的组合数目会更多,即选取出来的成像参数会更合理。比如,曝光时间的初始值为3,结束值为5,步长为1,那么曝光时间可以是3、4、5;增益的初始值为4,结束值为8,步长为2,那么增益可以是4、6、8,即成像参数组合就有3x3=9种。选定任意一组成像参数(包括曝光和增益),拍摄待检测物体,在该图像中标示出m个感兴趣区域,并给定待搜索的成像参数的范围,即曝光和增益等的初始值、结束值以及步长,以及要设定的成像参数数目n。可以理解的是,m、n的取值为大于等于1的整数。遍历提供的曝光时间和增益成像参数的组合,在不同成像参数下拍摄多张图像以获取多张不同成像参数图像所对应的感兴趣区域。利用特征检测算法提取出每个感兴趣区域的特征,并计算出不同成像参数图像中每个感兴趣区域的特征点数目,并基于每个感兴趣区域的特征点数目的多少对成像参数进行排序,得到每个感兴趣区域在每个成像参数下的排序结果;以及,根据预先设置的成像参数数目n筛选出n组成像参数组成的成像参数组合。这个成像参数组合可以认为是成像质量最优的成像参数组合。
在一个实施例中,感兴趣区域可以通过手动框出,也可以通过算法推荐。
在一个实施例中,获取不同成像参数图像中每个感兴趣区域的特征点数目可以采用尺度不变性特征变换(Scale-invariant feature transform,SIFT)、加速鲁棒性特征(Speeded Up Robust Features,SURF)等特征检测算法,在此对获取特征点数目的方法不做任何的限制。
在一个实施例中,可以通过一种投票法筛选出n组成像参数组成的成像参数组合。根据排序结果统计出每个名次下每个感兴趣区域的成像参数,并统计每个名次下每个成像参数的频数,并按照从第一名到最后一名重要性逐渐降低的方式,对频数进行从高到低地排序,得到频数的排序;若没有平数,则按照频数的排序从高到低的选取预先设置的成像参数数目的成像参数组成成像参数组合;
若有平数,则根据频数的排序中出现平数的后面一名到最后一名重要性逐渐降低的方式依次进行重新排序,直至选出频数高的所述预先设置的成像参数数目的成像参数组成所述成像参数组合。比如第一名出现平数,则根据成像参数在第二名次下的频数,并选取在第二名次中频数最大的成像参数,如果比较第二名次下的频数时再次出现平票,继续比较平票成像参数在第三名次的频数,直至能区分出平票成像参数的名次,由此选出最佳的n组成像参数,并组成最佳成像参数集合。
请参考表1,表1中有9组成像参数,6个感兴趣区域,第1个感兴趣区域在第4组成像参数下所获取图像的特征点数目最多,即第4组成像参数为第1个感兴趣区域的第一名。同样的,第4组成像参数为第2个感兴趣区域的第一名,第9组成像参数为第3个感兴趣区域的第一名,第4组成像参数为第4个感兴趣区域的第一名,第6组成像参数为第5个感兴趣区域的第一名,第9组成像参数为第6个感兴趣区域的第一名。
根据统票结果:第4组成像参数为3个感兴趣区域的第一名,第9组成像参数为2个感兴趣区域的第一名,第6组成像参数为1个感兴趣区域的第一名。根据需求可以选择筛选出n组成像参数,若n=1,即选择第4组成像参数,如此,使用第4组成像参数拍摄一张图像,即适合统计所有感兴趣区域成像信息;若n=2,即选择第4组和第9组成像参数,使用第4组成像参数拍摄第一张图像,适合统计该图像中的第1个、第2个、第4个感兴趣区域成像信息,使用第9组成像参数拍摄第二张图像,适合统计该图像中的第3个、第6个感兴趣区域成像信息,而第5个感兴趣区域由于在第4组成像参数的特征点数目更多,即选择使用第一张图像来统计第5个感兴趣区域的成像信息。如此,能对不同的感兴趣区域在适合的成像参数下统计成像信息,从而提高感兴趣区域的成像质量。
可以理解的是,上述只是对n=1和n=2进行说明,n也可以根据需求按上述实施例的方法筛选成像参数组合。
表1 成像参数及感兴趣区域的对照图
Figure PCTCN2020077859-appb-000001
Figure PCTCN2020077859-appb-000002
在一些情况下,同一名次下的成像参数可能出现平票,则需要比较平票的成像参数在下一个排名出现的频数,直至能区分出平票成像参数的名次。请参考表2,表2中有9组成像参数,6个感兴趣区域,第1个感兴趣区域在第4组成像参数下所获取图像的特征点数目最多,即第4组成像参数为第1个感兴趣区域的第一名。同样的,第4组成像参数为第2个感兴趣区域的第一名,第9组成像参数为第3个感兴趣区域的第一名,第1组成像参数为第4个感兴趣区域的第一名,第6组成像参数为第5个感兴趣区域的第一名,第9组成像参数为第6个感兴趣区域的第一名。
根据统票结果:第4组成像参数为2个感兴趣区域的第一名,第9组成像参数为2个感兴趣区域的第一名,第1组成像参数为1个感兴趣区域的第一名,第6组成像参数为1个感兴趣区域的第一名。根据需求可以选择筛选出n组成像参数,若n=1,由于第4组和第9组成像参数都对应2个感兴趣区域,因此需要统计第4组和第9组成像参数在第二名次出现的频数,第3组成像参数为第1个感兴趣区域的第二名,第5组成像参数为第2个感兴趣区域的第二名,第8组成像参数为第3个感兴趣区域的第二名,第5组成像参数为第4个感兴趣区域的第二名,第7组成像参数为第5个感兴趣区域的第二名,第1组成像参数为第6个感兴趣区域的第二名。由于并未出现第4组和第9组成像参数,因此依旧是平票状态,那么需要统计第4组和第9组成像参数在第三名次出现的频数。
统计各感兴趣区域对第4组和第9组成像参数在第三名次出现的频数,第5组成像参数为第1个感兴趣区域的第三名,第6组成像参数为第2个感兴趣区域的第三名,第7组成像参数为第3个感兴趣区域的第三名,第7组成像参数为第4个感兴趣区域的第三名,第4组成像参数为第5个感兴趣区域的第三名,第8组成像参数为第6个感兴趣区域的第三名。因为第4组成像参数为第5个感兴趣区域的第三名,而第9组成像参数并未出现,那么第4组成像参数为第一名。
因此,若n=1,即选择第4组成像参数,那么使用第4组成像参数拍摄一张图像,即适合统计所有感兴趣区域成像信息;若n=2,即选择第4组和第9组成像参数,使用第4组成像参数拍摄第一张图像,适合统计该图像中的第1个、第2个、第4个感兴趣区域成像信息,使用第9组成像参数拍摄第二张图像,适合统计该图像中的第3个、第6个感兴趣区域成像信息,而第5个感兴趣区域由于在第4组成像参数的特征点数目更多,即选择使用第一张图像来统计第5个感兴趣区域的成像信息。能对不同的感兴趣区域在适合的成像参数下统计成像信息,从而提高感兴趣区域的成像质量。
表2 成像参数及感兴趣区域的对照图
Figure PCTCN2020077859-appb-000003
在另一个实施例中,可以通过另一种投票法筛选出n组成像参数。根据所述排序结果获取到每个感兴趣区域前k组的成像参数,其中,k≥2;统计出每个名次下每个感兴趣区域的成像参数,并统计每个名次下每个成像参数的频数,并按照从第一名到最后一名重要性逐渐降低的方式,对频数进行从高到低地排序,得到频数的排序;若没有平数,则按照频数的排序从高到低的选取所述预先设置的成像参数数目的成像参数组成所述成像参数组合;若有平数,则根据频数的排序中出现平数的后面一名到最后一名重要性逐渐降低的方式依次进行重新排序,直至选出频数高的所述预先设置的成像参数数目的成像参数组成所述成像参数组合。
请继续参考表1,对于9组成像参数,6个感兴趣区域,根据特征点数目的排序获取每个感兴趣区域的特征点数目最多的前K组的成像参数(K大于等于 2),也就是说,6个感兴趣区域可以对9组成像参数进行投票,且每个感兴趣区域可以投2票,即总共有12票。第1个感兴趣区域在第3组和第4组成像参数下所获取图像的特征点数目最多,即第1个感兴趣区域会把票投给第3组和第4组。同样的,第2个感兴趣区域会把票投给第4组和第5组,第3个感兴趣区域会把票投给第8组和第9组,第4个感兴趣区域会把票投给第4组和第5组,第5个感兴趣区域会把票投给第6组和第7组,第6个感兴趣区域会把票投给第9组和第8组。
根据统票结果:第4组成像参数获得3票,第5组、第8组和第9组成像参数各获得2票、第3组、第7组和第8组成像参数各获得1票。根据需求可以选择筛选出n组成像参数,若n=1,即选择第4组成像参数,如此,使用第4组成像参数拍摄第一张图像,适合统计所有感兴趣区域成像信息;若n=2,则会出现平票情况,那么需要在第5组、第8组和第9组之间选择第二名的成像参数。根据各感兴趣区域分别在第5组、第8组和第9组成像参数下所获取的特征点数目对第5组、第8组和第9组进行投票,那么第1个感兴趣区域会将票投给第5组,第2个感兴趣区域会将票投给第5组,第3个感兴趣区域会将票投给第9组,第4个感兴趣区域会将票投给第5组,第5个感兴趣区域会将票投给第5组,第6个感兴趣区域会将票投给第9组。那么第5组获得4票,第9组获得2票。
因此,若n=2,那么选择第4组和第5组成像参数,使用第4组成像参数拍摄第一张图像,适合统计该图像中的第1个、第2个、第4个感兴趣区域成像信息,使用第5组成像参数拍摄第二张图像,适合统计该图像中的第3个、第6个感兴趣区域成像信息,而第5个感兴趣区域由于在第4组成像参数的特征点数目更多,即选择使用第一张图像来统计第5个感兴趣区域的成像信息。如此,能对不同的感兴趣区域在适合的成像参数下统计成像信息,从而提高感兴趣区域的成像质量。
可以理解的是,n的值是根据用户的需求设定的,可以选择1组成像参数拍摄,也可以选择多组成像参数拍摄。用户可以根据自己的感兴趣区域选择适合的成像参数进行拍摄。
图2为本发明一实施例中一种设定成像参数的方法流程图,包括如下步骤:
S1:遍历获取相机在不同成像参数下的包含多个感兴趣区域的图像并提取每个图像中所述感兴趣区域的特征点数目;
具体地,相机11拍摄多张不同成像参数下包含多个感兴趣区域的图像,并将图像传至处理器12,处理器12接收该图像并利用特征检测算法提取图像的感兴趣区域的特征点数目。
S2:根据所述感兴趣区域的特征点数目的多少对每个所述成像参数进行排序,得到每个所述感兴趣区域在每个所述成像参数下的排序结果;
具体地,处理器12根据感兴趣区域的特征点数目对每个感兴趣区域的成像参数进行排序,统计每个成像参数在各名次下出现的频数。
S3:根据所述排序结果和预先设置的成像参数数目筛选出成像参数组合;
具体地,根据预先设置的成像参数数目n,将成像参数出现的频数从高到低进行排序并筛选出成像参数组合。
S4:确定所述成像参数组合中每一个成像参数对应检测的所述感兴趣区域。
具体地,若n=1,则使用单组成像参数检测所有感兴趣区域;若n大于1,则成像参数组合包括不止一组的成像参数,在不同成像参数下统计不同感兴趣区域的成像信息,根据各感兴趣区域在n组成像参数的特征点数目选择适合的成像参数拍摄。若感兴趣区域的排序结果中排名第一的成像参数在成像参数组合中,则选择排名第一的成像参数检测所述感兴趣区域;若感兴趣区域的排名第一的成像参数不在成像参数组合中,则按照排序结果中的顺序靠前的成像参数检测感兴趣区域,也可以理解为选择特征数目最多的成像参数作为检测该感兴趣区域的成像参数。
本发明实现上述实施例方法中的全部或部分流程,也可以通过计算机程序来指令相关的硬件来完成,所述的计算机程序可存储于一计算机可读存储介质中,该计算机程序在被处理器执行时,可实现上述各个方法实施例的步骤。其中,所述计算机程序包括计算机程序代码,所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括:能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、电载波信号、电信信号以及软件分发 介质等。需要说明的是,所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在某些司法管辖区,根据立法和专利实践,计算机可读介质不包括电载波信号和电信信号。
本发明根据图像中感兴趣区域特征点数目的多少来评估图像成像质量,并给出适合所有感兴趣区域的最佳成像参数,也可以给出多组成像参数以分别适合不同感兴趣区域。与人眼视觉认知系统类似,一般光照环境越好,人眼可观察到的目标的细节越多,而图像中提取到的特征点数目越多,说明图像的光照环境越好,所以根据特征点数目多少推荐的最佳成像参数下的图像,与人视觉上选取的优质图像是一致的。
如图3所示,利用本发明提供的系统或方法,用户根据实际需求,选定任意一组成像参数(包括曝光和增益),拍摄待检测物体,在该图像中选定感兴趣区域,并给定待搜索的成像参数的范围,即曝光和增益等的初始值、结束值以及步长,以及要设定的成像参数数目;当相机光圈、焦距调节至合适值后,相机自动拍摄一系列成像参数(曝光、增益)组合下的图像,通过计算所获取感兴趣图像区域的特征点数目,基于特征点数目对成像参数进行排序,采用投票法推荐出适合感兴趣区域的合理成像参数组合,用户可根据推荐的成像参数组合进一步确认选取,然后拍摄所需图像。
以上内容是结合具体的优选实施方式对本发明所做的进一步详细说明,不能认定本发明的具体实施只局限于这些说明。对于本发明所属技术领域的技术人员来说,在不脱离本发明构思的前提下,还可以做出若干等同替代或明显变型,而且性能或用途相同,都应当视为属于本发明的保护范围。

Claims (10)

  1. 一种设定成像参数的方法,其特征在于,包括如下步骤:
    S1:遍历获取相机在不同成像参数下的包含多个感兴趣区域的图像并提取每个图像中所述感兴趣区域的特征点数目;
    S2:根据所述感兴趣区域的特征点数目的多少对每个所述成像参数进行排序,得到每个所述感兴趣区域在每个所述成像参数下的排序结果;
    S3:根据所述排序结果和预先设置的成像参数数目筛选出成像参数组合;
    S4:确定所述成像参数组合中每一个成像参数对应检测的所述感兴趣区域。
  2. 如权利要求1所述的设定成像参数的方法,其特征在于,采用投票法根据所述排序结果和所述预先设置的成像参数数目筛选出成像参数组合。
  3. 如权利要求2所述的设定成像参数的方法,其特征在于,筛选出成像参数组合包括如下步骤:
    根据所述排序结果统计出每个名次下每个所述感兴趣区域的成像参数,并统计每个所述名次下每个所述成像参数的频数,并按照从第一名到最后一名重要性逐渐降低的方式,对频数进行从高到低地排序,得到频数的排序;
    若没有平数,则按照所述频数的排序从高到低的选取所述预先设置的成像参数数目的成像参数组成所述成像参数组合;
    若有平数,则根据所述频数的排序中出现平数的后面一名到最后一名重要性逐渐降低的方式依次进行重新排序,直至选出频数高的所述预先设置的成像参数数目的成像参数组成所述成像参数组合。
  4. 如权利要求2所述的设定成像参数的方法,其特征在于,筛选出成像参数组合包括如下步骤:
    根据所述排序结果获取到每个所述感兴趣区域前k组的成像参数,其中,k≥2;
    统计出每个名次下每个所述感兴趣区域的成像参数,并统计每个所述名次下每个所述成像参数的频数,并按照从第一名到最后一名重要性逐渐降低的方式,对频数进行从高到低地排序,得到频数的排序;
    若没有平数,则按照所述频数的排序从高到低的选取所述预先设置的成像参数数目的成像参数组成所述成像参数组合;
    若有平数,则根据所述频数的排序中出现平数的后面一名到最后一名重要性 逐渐降低的方式依次进行重新排序,直至选出频数高的所述预先设置的成像参数数目的成像参数组成所述成像参数组合。
  5. 如权利要求1-4任一所述的设定成像参数的方法,其特征在于,确定所述成像参数组合中每一个所述成像参数对应检测的所述感兴趣区域包括:
    若所述感兴趣区域的所述排序结果中排名第一的成像参数在所述成像参数组合中,则选择所述排名第一的成像参数检测所述感兴趣区域;
    若所述感兴趣区域的排名第一的成像参数不在所述成像参数组合中,则按照所述排序结果中的顺序靠前的所述成像参数检测所述感兴趣区域。
  6. 如权利要求1-4任一所述的设定成像参数的方法,其特征在于,所述成像参数的范围根据曝光和增益的初始值、结束值以及步长确定。
  7. 如权利要求1-4任一所述的设定成像参数的方法,其特征在于,通过算法框出所述感兴趣区域。
  8. 如权利要求1-4任一所述的设定成像参数的方法,其特征在于,采用包括尺度不变性特征变换、加速鲁棒性特征的特征检测算法获取所述感兴趣区域的特征点数目。
  9. 一种设定成像参数系统,其特征在于,包括:
    相机,用于获取在不同成像参数下的包含多个感兴趣区域的图像,并将所述图像传至处理器;
    所述处理器,用于执行以下功能:
    接收所述图像并提取每个所述感兴趣区域的特征点数目;
    根据所述感兴趣区域的特征点数目的多少对每个所述成像参数进行排序,得到每个所述感兴趣区域在每个所述成像参数下的排序结果;
    根据所述排序结果和预先设置的成像参数数目筛选出成像参数组合;
    确定所述成像参数组合中每一个成像参数对应检测的所述感兴趣区域。
  10. 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,其特征在于,所述计算机程序被处理器执行时实现如权利要求1-8任一所述方法的步骤。
PCT/CN2020/077859 2020-01-10 2020-03-04 一种设定成像参数的方法、系统及计算机可读存储介质 WO2021138988A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010027438.4A CN111212223B (zh) 2020-01-10 2020-01-10 一种设定成像参数的方法、系统及计算机可读存储介质
CN202010027438.4 2020-01-10

Publications (1)

Publication Number Publication Date
WO2021138988A1 true WO2021138988A1 (zh) 2021-07-15

Family

ID=70790024

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/077859 WO2021138988A1 (zh) 2020-01-10 2020-03-04 一种设定成像参数的方法、系统及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN111212223B (zh)
WO (1) WO2021138988A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104463846A (zh) * 2014-11-04 2015-03-25 浙江捷尚视觉科技股份有限公司 一种用于数字图像处理中的参数调整方法
CN104937637A (zh) * 2013-07-31 2015-09-23 Plk科技株式会社 用于识别交通标志牌的车辆用影像识别系统
CN104997533A (zh) * 2015-06-23 2015-10-28 武汉超信电子工程有限公司 超声探头几何参数自动校正方法和装置
US20170059812A1 (en) * 2015-08-25 2017-03-02 Electronics And Telecommunications Research Institute Imaging apparatus and method for operating the same
CN108322666A (zh) * 2018-02-12 2018-07-24 广州视源电子科技股份有限公司 摄像头快门的调控方法、装置、计算机设备及存储介质
CN109194868A (zh) * 2018-09-03 2019-01-11 信利光电股份有限公司 一种图像优化方法、装置、设备及存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5544764B2 (ja) * 2009-06-09 2014-07-09 ソニー株式会社 画像処理装置および方法、並びにプログラム
CN103020947B (zh) * 2011-09-23 2016-04-06 阿里巴巴集团控股有限公司 一种图像的质量分析方法及装置
CN108346139A (zh) * 2018-01-09 2018-07-31 阿里巴巴集团控股有限公司 一种图像筛选方法及装置
CN110097001A (zh) * 2019-04-30 2019-08-06 恒睿(重庆)人工智能技术研究院有限公司 生成最佳多人脸图像的方法、系统、设备及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104937637A (zh) * 2013-07-31 2015-09-23 Plk科技株式会社 用于识别交通标志牌的车辆用影像识别系统
CN104463846A (zh) * 2014-11-04 2015-03-25 浙江捷尚视觉科技股份有限公司 一种用于数字图像处理中的参数调整方法
CN104997533A (zh) * 2015-06-23 2015-10-28 武汉超信电子工程有限公司 超声探头几何参数自动校正方法和装置
US20170059812A1 (en) * 2015-08-25 2017-03-02 Electronics And Telecommunications Research Institute Imaging apparatus and method for operating the same
CN108322666A (zh) * 2018-02-12 2018-07-24 广州视源电子科技股份有限公司 摄像头快门的调控方法、装置、计算机设备及存储介质
CN109194868A (zh) * 2018-09-03 2019-01-11 信利光电股份有限公司 一种图像优化方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN111212223B (zh) 2021-01-22
CN111212223A (zh) 2020-05-29

Similar Documents

Publication Publication Date Title
US8009208B2 (en) Detection and removal of blemishes in digital images utilizing original images of defocused scenes
US8957993B2 (en) Detecting red eye filter and apparatus using meta-data
US7424170B2 (en) Automated statistical self-calibrating detection and removal of blemishes in digital images based on determining probabilities based on image analysis of single images
US7369712B2 (en) Automated statistical self-calibrating detection and removal of blemishes in digital images based on multiple occurrences of dust in images
US7206461B2 (en) Digital image acquisition and processing system
CN101867724B (zh) 成像装置以及成像方法
TWI389559B (zh) 前景影像分離方法
US20050031224A1 (en) Detecting red eye filter and apparatus using meta-data
US7308156B2 (en) Automated statistical self-calibrating detection and removal of blemishes in digital images based on a dust map developed from actual image data
US20050068450A1 (en) Automated statistical self-calibrating detection and removal of blemishes in digital images dependent upon changes in extracted parameter values
JP2007507932A (ja) 補正方法
WO2019210707A1 (zh) 一种图像清晰度评测方法、装置及电子设备
CN110248170A (zh) 图像色彩调整方法及装置
CN111656781A (zh) 用于使用参考图像进行图像信号处理器调谐的系统和方法
CN112188190A (zh) 污渍检测方法、烹饪器具、服务器和存储介质
JP2018107593A (ja) 画像処理装置、画像処理方法およびプログラム
CN111630854A (zh) 用于图像信号处理器调谐的系统和方法
JP2004005384A (ja) 画像処理方法、画像処理装置、プログラム及び記録媒体、自動トリミング装置、並びに肖像写真撮影装置
WO2021138988A1 (zh) 一种设定成像参数的方法、系统及计算机可读存储介质
CN105260997A (zh) 一种自动获取目标图像的方法
JP2010035177A (ja) 画像撮影装置及び画像処理システム
CN111615827A (zh) 用于图像信号处理器调谐的系统和方法
CN107995476B (zh) 一种图像处理方法及装置
CN110460773B (zh) 图像处理方法和装置、电子设备、计算机可读存储介质
CN113840080A (zh) 摄像设备、摄像设备的控制方法和计算机可读介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20912044

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20912044

Country of ref document: EP

Kind code of ref document: A1