WO2016127478A1 - 一种图像处理方法、装置和终端 - Google Patents
一种图像处理方法、装置和终端 Download PDFInfo
- Publication number
- WO2016127478A1 WO2016127478A1 PCT/CN2015/075563 CN2015075563W WO2016127478A1 WO 2016127478 A1 WO2016127478 A1 WO 2016127478A1 CN 2015075563 W CN2015075563 W CN 2015075563W WO 2016127478 A1 WO2016127478 A1 WO 2016127478A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- images
- matching degree
- preset
- threshold
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
Definitions
- the present invention relates to the field of image processing technologies, and in particular, to an image processing method, apparatus, and terminal.
- the functions in the terminals are becoming more and more abundant.
- the user uses a relatively large number of functions as a photographing function.
- the user often corrects the same scene due to jitter, focus failure, other people walking through the framing frame, adjusting camera parameters, not taking a good posture, too dark light, too much noise, and the like.
- the photographs shown in FIG. 1B are obviously blurred; the image shown in FIG. 1C is relatively clear; the image shown in FIG. 1A is smaller in pixels.
- the display on the terminal screen appears to be clear, but when displayed on a higher-pixel terminal screen, the image is blurred and more noisy.
- the terminal's album does not have an automatic management function for this situation, that is, the same photos of the scene cannot be uniformly provided to the user, and the photo deletion or retention management is performed, but the user needs to manually organize the shooting in the repeated scenes in the terminal. Failed photos, therefore, the user is very inconvenient when managing the same photos of the scene, which is easy to cause a problem in the terminal to store a large number of photographs that failed to be photographed, wasting the storage space of the terminal.
- a main object of the present invention is to provide an image processing method, apparatus, and terminal to solve the problem of inconvenient management of the same scene in the prior art.
- the present invention provides an image processing method including: in a plurality of images to be processed, according to each Feature of the image, calculating a matching degree between the images; acquiring a plurality of images whose matching degree is greater than a preset matching degree threshold; deleting or retaining the matching degree greater than a preset matching degree threshold according to a preset image management policy All or part of the image in multiple images.
- calculating the matching degree between the images according to the characteristics of each image including: dividing, in the image to be processed, the image to be processed according to an attribute of each image And a plurality of image groups; respectively calculating a matching degree between the images in each image group; and obtaining a plurality of images whose matching degree is greater than a preset matching degree threshold, comprising: respectively obtaining matching in each image group Multiple images with a degree greater than the preset match threshold.
- the attributes of the image include: an exchangeable image file EXIF attribute; wherein the EXIF attribute includes image capturing time information and image capturing position information; and the plurality of images to be processed are divided into one according to attributes of each image Or a plurality of image groups, including: dividing, according to the EXIF attribute of each image, a plurality of images whose shooting distance is less than a preset distance threshold and the shooting time interval is less than a preset time threshold, into one image group.
- the calculating the degree of matching between the images in each image group separately comprises: obtaining, for each image group, a thumbnail of each image in the image group; extracting, in each image thumbnail Visual image feature points; based on the visual image feature points of each image, the degree of matching between the two images in the image group is calculated.
- the deleting or retaining all or part of the plurality of images with the matching degree greater than the preset matching degree threshold according to the preset image management policy including: the matching degree is greater than the preset matching degree a plurality of images of the threshold, calculating the sharpness and/or the number of noises of each of the images; selecting to delete or retain all or part of the plurality of images according to the sharpness and/or the number of noises of each image .
- the present invention further provides an image processing apparatus, comprising: a calculation module, configured to calculate a degree of matching between images to be processed; and an acquisition module, configured to acquire a plurality of images whose matching degree is greater than a preset matching degree threshold; And a module, configured to delete or retain all or part of the plurality of images whose matching degree is greater than a preset matching degree threshold according to a preset image management policy.
- a calculation module configured to calculate a degree of matching between images to be processed
- an acquisition module configured to acquire a plurality of images whose matching degree is greater than a preset matching degree threshold
- a module configured to delete or retain all or part of the plurality of images whose matching degree is greater than a preset matching degree threshold according to a preset image management policy.
- the calculation module is configured to divide the image to be processed into one or more image groups according to attributes of each image in an image to be processed; respectively calculate between images in each image group The matching degree; wherein the attributes of the image include: an exchangeable image file EXIF attribute; wherein The EXIF attribute includes image capturing time information and image capturing position information; the calculating module, according to the EXIF attribute of each image, the shooting distance is less than a preset distance threshold, and the shooting time interval is less than a preset time threshold.
- the image is divided into an image group; the acquiring module is configured to respectively acquire a plurality of images in each image group whose matching degree is greater than a preset matching degree threshold.
- the calculation module is configured to: obtain, for each image group, a thumbnail of each image in the image group; extract a visual image feature point in a thumbnail of each image; and visualize each image based on each image Image feature points, calculating the degree of matching between the two images in the image group.
- the processing module is configured to: calculate, for the plurality of images whose matching degree is greater than a preset matching degree threshold, the number of sharpness and/or noise of each image; according to the clarity of each image / or the number of noises, choose to delete or retain all or part of the images.
- the present invention further provides a terminal that performs image processing on an image stored in the terminal based on the image processing apparatus.
- the invention is based on image recognition technology and signal processing technology, and recognizes photos of the same scene in multiple photos, performs image deletion or retention management; can also calculate the sharpness of photos of the same scene, and clear the photos of the same scene
- the degree is displayed to the user as a reference when the user organizes the photo.
- the invention can help the user to recognize repetitive photos, and is convenient for the user to delete the photographs that failed to be photographed, thereby saving storage space and having a better user experience.
- 1A, 1B and 1C are schematic views of photographs taken repeatedly in the same scene
- FIG. 2 is a flow chart of an image processing method according to an embodiment of the present invention.
- FIG. 3 is a specific flowchart of an image processing method according to an embodiment of the present invention.
- FIG. 4 is a flow chart showing the steps of calculating the degree of matching between images in accordance with an embodiment of the present invention
- FIG. 5 is a structural diagram of an image processing apparatus according to an embodiment of the present invention.
- FIG. 6 is a structural diagram of a terminal according to an embodiment of the present invention.
- the main idea of the present invention is to identify, in the image to be processed, a plurality of images in the same scene by recognizing the degree of matching between the images, thereby deleting or retaining all or part of the images in the plurality of images.
- the invention can help the user to recognize repetitive photos, and is convenient for the user to delete the photographs that failed to be photographed, thereby saving storage space and having a better user experience.
- the present invention provides an image processing method.
- 2 is a flow chart of an image processing method according to an embodiment of the present invention.
- Step S210 in the plurality of images to be processed, the degree of matching between the images is calculated according to the characteristics of each image.
- a plurality of images to be processed are, for example, photos and pictures.
- the degree of matching refers to the degree of similarity between images. Since the photos taken in the same scene are necessarily similar in height, the matching degree between the images can be used to filter out images of the same scene.
- all or part of the image is acquired as an image to be processed from the already stored image; the number of images to be processed is plural, at least two; and features in each image are extracted, the feature includes points and lines Or surface features, for example: features are visual image feature points; according to features in the image, the degree of matching between any plurality of images is calculated, for example, calculating the matching degree between the two images to be processed, and calculating the to-be-processed The degree of matching between any three images in the image. Further, based on the preset matching degree algorithm, the degree of matching between the images is calculated.
- the matching algorithm includes but is not limited to: Euclidean distance algorithm and cosine theorem algorithm.
- Step S220 Acquire a plurality of images whose matching degree is greater than a preset matching degree threshold.
- the match threshold is used to measure whether the images are the same scene. If the matching degree between the images is greater than the matching degree threshold, it means that the scenes between the images are the same; if the matching degree between the images is less than or equal to the matching degree threshold, it means that the scenes between the images are different.
- this embodiment Because the purpose of this embodiment is to manage images in the same scene. Therefore, after obtaining the matching degree between the images, a plurality of images whose matching degree is greater than the matching degree threshold are separately obtained, that is, a plurality of images of the same scene are separately acquired.
- the image to be processed includes a, b, c, d; the matching threshold is 0.9; the degree of matching between a and b, a and c, a and d, b and c, b and d, c and d is calculated.
- Step S230 deleting or retaining all or part of the plurality of images whose matching degree is greater than a preset matching degree threshold according to a preset image management policy.
- a processing method for a plurality of images whose matching degree is greater than a preset matching degree threshold is specified to delete or retain an image in the same scene.
- the image management strategy includes, but is not limited to, calculating the sharpness and/or the number of noises of each image for a plurality of images whose matching degree is greater than a preset matching degree threshold; according to the sharpness and/or noise of each image Number, select to delete or retain all or part of the images in the plurality of images.
- the image management strategy is: deleting, in a plurality of images whose matching degree is greater than a preset matching degree threshold, an image whose resolution is smaller than a preset sharpness threshold; or deleting an image whose noise number is greater than a threshold number of noise; or Sort the sharpness or noise of the picture, retain only the image with the highest resolution or the least number of noises, delete the remaining images; or, display multiple images with the matching degree greater than the matching threshold to the user. Choose to delete or keep the image.
- the definition of each image in the plurality of images whose matching degree is greater than the preset matching degree threshold is calculated; while the image is displayed, the sharpness of the image is displayed; the user can display the image according to the image The clarity to choose to delete or retain the image.
- each set of images is respectively presented to the user. For example, if the matching degree of a and b is greater than the matching degree threshold, and the matching degree of c and d is greater than the matching degree threshold, then a and b can be displayed for the user first, and after the processing of a and b is completed, the user is displayed c and d. .
- FIG. 3 is a specific flowchart of an image processing method according to an embodiment of the present invention.
- Step S310 in the stored image, acquiring an image that has not been processed as a to-be-processed image image.
- the marked processed image that is, after the image processing, is marked for the retained image to distinguish between the processed and unprocessed images.
- the mark of the image is read, and the unmarked image is acquired as the image to be processed.
- Step S320 in the image to be processed, the image to be processed is divided into one or more image groups according to the attributes of each image.
- the attributes of the image include: an Exchangeable Image File (EXIF) attribute; wherein, the EXIF attribute includes: image capturing time information, image capturing position information.
- the image capturing time information may be the time at which the image was taken.
- the image capturing location information may be positioning information of a Global Positioning System (GPS).
- GPS Global Positioning System
- the shooting time of the image is compared, the shooting positions of the images are compared, and the image whose shooting distance is less than the preset distance threshold and the shooting time interval is less than the preset time threshold is divided into An image group.
- the purpose of this is to divide an image with a close shooting distance and a close shooting time into one image group. Because the image is close in distance and the shooting time is close, the scene is more likely to be the same.
- the image to be processed is divided into a plurality of image groups, and then processed separately for each image group, thereby reducing the amount of calculation, improving the operation efficiency, and improving the extraction efficiency of the image in the same scene.
- the distance threshold is 100 m and the time threshold is 3 min. That is to say, the distance between the photographing locations of the images is less than 100 m, and the images with the photographing time interval of less than 3 min can be divided into one image group.
- step S330 the degree of matching between the images in each image group is calculated separately.
- each image group the following steps are respectively performed: acquiring thumbnails of each image in the image group; extracting visual image feature points in thumbnails of each image; calculating images based on visual image feature points of each image The degree of matching of any two images in the group.
- steps are respectively performed: acquiring thumbnails of each image in the image group; extracting visual image feature points in thumbnails of each image; calculating images based on visual image feature points of each image The degree of matching of any two images in the group.
- Step S340 respectively acquiring a plurality of images in each image group whose matching degree is greater than a matching degree threshold as a matching image.
- a plurality of images whose matching degree is greater than the matching degree threshold are mutually referred to as matching images.
- the scenes that can be called mutually matching images are the same scene.
- the image group contains images a, b, c, d; the matching degree of a and b is greater than the matching degree threshold; the matching degree of c and d is greater than the matching degree threshold; only a and b match each other, only c and d mutually matches the matching image.
- step S350 the sharpness of each matching image in each image group is calculated separately.
- the sharpness of the image is calculated based on the change in the frequency of the image signal in the image.
- the sharpness of each matched image can be quantified.
- the total range of sharpness is 0%-100%, increasing from 0%-100% sharpness, 0% is unclear, 100% is the clearest; the sharpness is divided into 5 levels; the first level is sharp The range is from 0% to 20%, the second level of resolution ranges from 21% to 40%, the third level of resolution ranges from 41% to 60%, and the fourth level of resolution ranges from 61% to 80%.
- the 5th level of resolution ranges from 81% to 100%.
- Step S360 respectively showing each matching image in each image group and its corresponding sharpness.
- each matching image in each image group and its corresponding sharpness level are displayed.
- an image group contains two sets of images with the same scene, then only one set of images with the same scene and their sharpness (or sharpness level) can be displayed at a time.
- Step S370 selecting to delete or retain the image according to each matching image and its corresponding sharpness.
- the sharpness threshold is 50%, and when the sharpness of the image is less than 50%, the image is directly deleted.
- sort the sharpness of the image select the image with the highest resolution, and delete other matching images.
- the user manually retains or deletes the image based on the matching image displayed and the clarity of the matching image.
- the number of noises of the image may be calculated according to the change of the frequency of the image signal in the image, and the number of noises of the image is displayed together with the image as reference data for image retention or deletion.
- step S330 Specifically for step S330,
- FIG. 4 is a flow chart of the steps of calculating the degree of matching between images, in accordance with an embodiment of the present invention.
- This embodiment introduces the Android (Android) version of OpenCV (Open Source Computer). Vision Library) function library.
- the degree of matching between images is calculated by the functions of Android and the functions in the OpenCV function library.
- Step S410 acquiring thumbnails of each image in the image group.
- the thumbnail of the image refers to a small image with a lower pixel obtained after the image is compressed.
- the ThumbnailUtils class is a function class for Android. For example: Get the thumbnail of the image with the following statement:
- Bitmap bitmap ThumbnailUtils.extractThumbnail(bitmap,width,height,ThumbnailUtils.OPTIONS_RECYCLE_INPUT)
- Step S420 extracting visual image feature points in the thumbnail of each image.
- a visual image feature point refers to a local feature of the image.
- the feature points of the visual image are: Scale-invariant feature transform (SIFT) feature points, or SURF (Speed Up robust feature, SURF) feature points.
- SIFT Scale-invariant feature transform
- SURF Speed Up robust feature, SURF
- the SIFT feature is a descriptor in the field of image processing. This descriptor has scale invariance and can detect key points in the image; SIFT is a 128-dimensional vector (feature vector).
- the visual image feature points can be extracted by the SIFT algorithm in the OpenCV function library, for example, the following image is used to extract the visual image feature points in the thumbnail:
- Step S430 calculating a matching degree between the two images in the image group based on the visual image feature points of each image.
- the matching degree of the two images is obtained by calculating the Euclidean distance of each visual image feature point in the two images. Because the SIFT feature is used to calculate the matching degree of the two images as the prior art, only one example is given here to make the embodiment more convenient to understand, and specific technical details are not described herein.
- the visual image feature point k1 in the first image For example, based on the visual image feature point k1 in the first image, find two visual image feature points (first key point and second key point) closest to the visual image feature point k1 in the second image, if The distance between the first key point and the visual image feature point k1 is less than or equal to the distance between the second key point and the visual image feature point k1, and the distance corresponding to the first key point is divided by the distance corresponding to the second key point, if the obtained value If the ratio is greater than the preset ratio, the first image and the second image match.
- the present invention also provides an image processing apparatus.
- the image processing apparatus may be disposed in the terminal, and the image processing apparatus is configured to perform image processing on the image stored in the terminal.
- FIG. 5 it is a structural diagram of an image processing apparatus according to an embodiment of the present invention.
- the device including:
- the calculating module 510 is configured to calculate a degree of matching between the images to be processed.
- the obtaining module 520 is configured to obtain multiple images whose matching degree is greater than a preset matching degree threshold.
- the processing module 530 is configured to delete or retain all or part of the plurality of images whose matching degree is greater than a preset matching degree threshold according to a preset image management policy.
- the calculating module 510 is configured to divide the image to be processed into one or more image groups according to attributes of each image in the image to be processed; respectively calculate each image group a degree of matching between the images; wherein the attributes of the image include: an exchangeable image file EXIF attribute; wherein the EXIF attribute includes image capturing time information and image capturing position information; and the calculating module is configured according to each image
- the EXIF attribute divides a plurality of images whose shooting distance is less than a preset distance threshold and the shooting time interval is less than a preset time threshold into one image group.
- the obtaining module 520 is configured to respectively acquire a plurality of images in each image group whose matching degree is greater than a preset matching degree threshold.
- the calculation module 510 is configured to acquire, for each image group, a thumbnail of each image in the image group; in the thumbnail of each image, extract a visual image feature point; and display a visual image based on each image Feature points, the degree of matching between the two images in the image group is calculated.
- the processing module 530 is configured to calculate a sharpness and/or a number of noises of each image for the plurality of images whose matching degree is greater than a preset matching degree threshold; The number of sharpness and/or noise is selected to delete or retain all or part of the plurality of images.
- the invention also provides a terminal.
- the terminal can include at least one input device 603, at least one output device 604, at least one processor 601, such as a CPU, a memory 605, and at least one bus 602.
- the bus 602 is used to connect the input device 603, the output device 604, the processor 601, and the memory 605.
- the above memory 605 may be a high speed RAM memory or a non-volatile memory such as a disk memory.
- the memory 605 is further configured to store a set of program codes, and the input device 603, the output device 604, and the processor 601 are configured to call the program code stored in the memory 605, and perform the following operations:
- the processor 601 is configured to calculate a matching degree between the images according to characteristics of each image in the plurality of images to be processed;
- the input device 603 is configured to acquire a plurality of images whose matching degree is greater than a preset matching degree threshold;
- the processor 601 is further configured to delete or retain all or part of the plurality of images whose matching degree is greater than a preset matching degree threshold by using the output device 604 according to a preset image management policy.
- the processor 601 calculates the matching degree between the images according to the characteristics of each image in the plurality of images to be processed, including:
- the processor 601 divides the image to be processed into one or more image groups according to attributes of each image in an image to be processed; and respectively calculates a matching degree between images in each image group;
- the attribute of the image includes: an exchangeable image file EXIF attribute; wherein the EXIF attribute includes image capturing time information and image capturing position information; and the processor 601 sets the shooting distance to be smaller according to an EXIF attribute of each image. a plurality of images of a preset distance threshold and a shooting time interval less than a preset time threshold are divided into one image group;
- the input device 603 respectively acquires a plurality of images in each image group whose matching degree is greater than a preset matching degree threshold.
- the processor 601 separately calculates a degree of matching between images in each image group, including:
- the input device 603 acquires a thumbnail of each image in the image group;
- the processor 601 extracts a visual image feature point in a thumbnail of each image
- the processor 601 calculates an image in the image group based on visual image feature points of each image The degree of matching between the two.
- the processor 601 deletes or retains all or part of the plurality of images whose matching degree is greater than a preset matching degree threshold by using the output device 604 according to a preset image management policy.
- the processor 601 calculates, for the plurality of images whose matching degree is greater than a preset matching degree threshold, the resolution and/or the number of noises of each image;
- the processor 601 selects to delete or retain all or part of the plurality of images by the output device 604 according to the sharpness and/or the number of noises of each image.
- the terminal introduced in the embodiment of the present invention may be used to implement some or all of the processes in the method embodiments introduced in conjunction with FIG. 2 to FIG.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Processing Or Creating Images (AREA)
Abstract
本发明公开了一种图像处理方法、装置和终端。该方法包括：在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度；获取匹配度大于预设的匹配度阈值的多个图像；根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。本发明基于图像识别技术和信号处理技术，在多张照片中，识别出相同场景的照片，进行图片删除或保留管理；还可以计算相同场景的照片的清晰度，将相同场景的照片及其清晰度展示给用户，作为用户整理照片时参考。通过本发明可以帮助用户识别重复性的照片，方便用户删除拍摄失败的照片，进而节省存储空间，用户体验效果较好。
Description
本申请要求于2015年2月15日提交中国专利局、申请号为201510082223.1，发明名称为“一种图像处理方法、装置和终端”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本发明涉及图像处理技术领域，尤其涉及一种图像处理方法、装置和终端。
随着终端技术的发展，终端中的功能越来越丰富。在终端的诸多功能中，用户使用比较多的功能为拍照功能。但是，在拍照过程中，用户经常由于抖动、对焦失败、取景框中有其他人走过、调节相机参数、拍照姿势没摆好、光线太暗、噪点过多等原因，对同一场景进行修正性的重复拍摄，在重复拍摄的照片中，大部分为拍摄失败的照片，只有少部分为拍摄成功的照片。
例如图1A、图1B和图1C所示的相同场景下的重复拍摄的照片，图1B所示的图像明显模糊；图1C所示的图像比较清晰；图1A所示的图像，在像素较小的终端屏幕上显示看似清晰，但是在像素较高的终端屏幕上显示就会发现图像模糊且噪点较多。
针对重复拍照的情况，终端中会存在多张场景相同的照片。目前，终端的图册没有针对这种情况的自动管理功能，即不能将场景相同的照片统一提供给用户，进行照片的删除或保留管理，而是需要用户自己手动整理终端中重复场景下拍摄的失败照片，因此，用户在管理场景相同的照片时非常不便，容易造成终端中存储大量拍摄失败的照片的问题，浪费终端的存储空间。
发明内容
本发明的主要目的在于提供一种图像处理方法、装置和终端，以解决现有技术对于场景相同的照片管理不便的问题。
基于上述技术问题，本发明是通过以下技术方案来解决的。
本发明提供了一种图像处理方法，包括：在待处理的多个图像中，根据每
个图像的特征，计算图像之间的匹配度；获取匹配度大于预设的匹配度阈值的多个图像；根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
其中，在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度，包括：在待处理的图像中，根据每个图像的属性，将所述待处理的图像划分为一个或多个图像组；分别计算每个图像组中的图像之间的匹配度；所述获取匹配度大于预设的匹配度阈值的多个图像，包括：分别获取每个图像组中匹配度大于预设的匹配度阈值的多个图像。
其中，所述图像的属性包括：可交换图像文件EXIF属性；其中，所述EXIF属性包括图像拍摄时间信息和图像拍摄位置信息；根据每个图像的属性，将待处理的多个图像划分为一个或多个图像组，包括：根据每个图像的EXIF属性，将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的多个图像，划分为一个图像组。
其中，所述分别计算每个图像组中的图像之间的匹配度，包括：针对每个图像组，获取所述图像组中每个图像的缩略图；在每个图像的缩略图中，提取视觉图像特征点；基于每个图像的视觉图像特征点，计算所述图像组中图像两两之间的匹配度。
其中，所述根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像，包括：针对所述匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；根据每个图像的清晰度和/或噪点个数，选择删除或保留所述多个图像中的全部或部分图像。
本发明还提供了一种图像处理装置，包括：计算模块，用于计算待处理的图像之间的匹配度；获取模块，用于获取匹配度大于预设的匹配度阈值的多个图像；处理模块，用于根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
其中，所述计算模块，用于在待处理的图像中，根据每个图像的属性，将所述待处理的图像划分为一个或多个图像组；分别计算每个图像组中的图像之间的匹配度；其中，所述图像的属性包括：可交换图像文件EXIF属性；其中，
所述EXIF属性包括图像拍摄时间信息和图像拍摄位置信息；所述计算模块，根据每个图像的EXIF属性，将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的多个图像，划分为一个图像组；所述获取模块，用于分别获取每个图像组中匹配度大于预设的匹配度阈值的多个图像。
其中，所述计算模块，用于：针对每个图像组，获取所述图像组中每个图像的缩略图；在每个图像的缩略图中，提取视觉图像特征点；基于每个图像的视觉图像特征点，计算所述图像组中图像两两之间的匹配度。
其中，所述处理模块，用于：针对所述匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；根据每个图像的清晰度和/或噪点个数，选择删除或保留所述多个图像中的全部或部分图像。
本发明又提供了一种终端，所述终端基于所述的图像处理装置，对存储在所述终端中的图像进行图像处理。
本发明有益效果如下：
本发明基于图像识别技术和信号处理技术，在多张照片中，识别出相同场景的照片，进行图片删除或保留管理；还可以计算相同场景的照片的清晰度，将相同场景的照片及其清晰度展示给用户，作为用户整理照片时参考。通过本发明可以帮助用户识别重复性的照片，方便用户删除拍摄失败的照片，进而节省存储空间，用户体验效果较好。
此处所说明的附图用来提供对本发明的进一步理解，构成本申请的一部分，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：
图1A、1B和1C是相同场景下重复拍摄的照片的示意图；
图2是根据本发明一实施例的图像处理方法的流程图；
图3是根据本发明一实施例的图像处理方法的具体流程图；
图4是根据本发明一实施例的计算图像之间的匹配度的步骤的流程图；
图5是根据本发明一实施例的图像处理装置的结构图；
图6是根据本发明一实施例的终端的结构图。
本发明的主要思想在于，在待处理的图像中，通过识别图像之间的匹配度，识别出相同场景下的多个图像，进而删除或保留多个图像中的全部或部分图像。通过本发明可以帮助用户识别重复性的照片，方便用户删除拍摄失败的照片，进而节省存储空间，用户体验效果较好。
为使本发明的目的、技术方案和优点更加清楚，以下结合附图及具体实施例，对本发明作进一步地详细说明。
本发明提供了一种图像处理方法。如图2所示，为根据本发明一实施例的图像处理方法的流程图。
步骤S210，在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度。
待处理的多个图像例如是照片、图片。
匹配度是指图像之间的相似程度。因为相同场景拍摄的照片必然相似度较高，所以可以用图像之间的匹配度来筛选出相同场景的图像。
具体的，从已经存储的图像中，获取全部或部分图像作为待处理的图像；待处理的图像的数量为多个，至少为两个；提取每个图像中的特征，该特征包括点、线或面特征，例如：特征为视觉图像特征点；根据图像中的特征，计算任意的多个图像之间的匹配度，如，计算待处理的图像两两之间的匹配度、计算待处理的图像之中任意三个图像之间的匹配度。进一步地，基于预设的匹配度算法，计算图像之间的匹配度。匹配度算法包括但不限于：欧几里得距离算法、余弦定理算法。
步骤S220，获取匹配度大于预设的匹配度阈值的多个图像。
匹配度阈值，用于衡量图像之间是否为相同场景。图像之间的匹配度大于匹配度阈值，则表示图像之间的场景相同；图像之间的匹配度小于等于匹配度阈值，则表示图像之间的场景不同。
因为，本实施例的目的在于，管理相同场景下的图像。所以，在得到图像之间的匹配度之后，将匹配度大于匹配度阈值的多个图像单独获取出来，也即是将相同场景的多个图像单独获取出来。
例如：待处理的图像包括a、b、c、d；匹配度阈值为0.9；计算a和b、a和c、a和d、b和c、b和d、c和d之间的匹配度分别为0.95(a和b)、0.2(a和c)、0.3(a和d)、0.2(b和c)、0.3(b和d)、0.96(c和d)；图像之间相似度大于0.9的为a和b、以及c和d，那么说明a和b为相同场景下的图像，c和d为相同场景下的图像；分别获取a和b、以及c和d，作为两组相同场景的图像。
步骤S230，根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
在图像管理策略中，规定了对匹配度大于预设的匹配度阈值的多个图像的处理方法，以便删除或保留相同场景下的图像。
图像管理策略包括但不限于：针对匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；根据每个图像的清晰度和/或噪点个数，选择删除或保留所述多个图像中的全部或部分图像。
例如：图像管理策略为：在匹配度大于预设的匹配度阈值的多个图像中，删除清晰度小于预设的清晰度阈值的图像；或者删除噪点个数大于噪点个数阈值的图像；或者对图片的清晰度或噪点个数进行排序，仅保留清晰度最高或噪点个数最少的图像，删除其余的图像；再或者，将匹配度大于匹配度阈值的多个图像展示给用户，由用户选择删除或保留图像。
为了方便用户选择需要删除或保留的图像，计算匹配度大于预设的匹配度阈值的多个图像中每个图像的清晰度；在展示图像的同时，展示该图像的清晰度；用户可以根据图像的清晰度来选择删除或保留该图像。
进一步地，如果有多组匹配度大于匹配度阈值的多个图像，也即是说，有多组相同场景的图像，则分别将每组图像展示给用户。例如：a和b的匹配度大于匹配度阈值，c和d的匹配度大于匹配度阈值，那么可以先为用户展示a和b，在对a和b处理完成后，再为用户展示c和d。
下面给出一个具体的实施例，来对本发明进行进一步地的说明。
如图3所示，图3是根据本发明一实施例的图像处理方法的具体流程图。
步骤S310，在存储的图像中，获取尚未被处理过的图像，作为待处理图
像。
在本实施例中，标记处理后的图像，也即是在图像处理后，针对被保留下来的图像，进行标记，以便区分经过处理和未经过处理的图像。
在存储的图像中，读取图像的标记，获取未被标记的图像，作为待处理图像。
步骤S320，在待处理的图像中，根据每个图像的属性，将待处理的图像划分为一个或多个图像组。
图像的属性包括：可交换图像文件(Exchangeable Image File，简称EXIF)属性；其中，EXIF属性包括：图像拍摄时间信息、图像拍摄位置信息。图像拍摄时间信息可以是拍摄图像的时刻。该图像拍摄位置信息可以是全球定位系统(Global Positioning System，简称GPS)的定位信息。
根据每个图像的EXIF属性，将图像的拍摄时间进行比较，将图像的拍摄位置进行比较，进而将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的图像，划分为一个图像组。这样的目的在于，可以将拍摄距离较近、且拍摄时间较近的图像划分到一个图像组。因为，拍摄距离近、且拍摄时间近的图像，场景相同的可能性更大。
进一步地，将待处理图像划分为多个图像组，进而针对每个图像组分别进行处理，这样可以降低运算量，提高运算效率，提高了同一场景下的图片的提取效率。
本实施例优选的，距离阈值为100m，时间阈值为3min。也即是说，图像之间的拍摄地点的距离小于100m，且拍摄的时间间隔小于3min的图像都可以划分到一个图像组中。
步骤S330，分别计算每个图像组中的图像之间的匹配度。
在每个图像组中，分别执行以下步骤：获取图像组中每个图像的缩略图；在每个图像的缩略图中，提取视觉图像特征点；基于每个图像的视觉图像特征点，计算图像组中任意两个图像的匹配度。上述步骤的详细描述请参照下面的图4。
步骤S340，分别获取每个图像组中匹配度大于匹配度阈值的多个图像，作为匹配图像。
匹配度大于匹配度阈值的多个图像之间，互相称作匹配图像。能够互相称作匹配图像的图片之间场景相同。例如：图像组中包含图像a、b、c、d；a和b匹配度大于匹配度阈值；c和d匹配度大于匹配度阈值；则仅有a和b互称匹配图像，仅有c和d互称匹配图像。
步骤S350，分别计算每个图像组中的每个匹配图像的清晰度。
根据图像中图像信号频率的变化，计算图像的清晰度。
进一步地，可以对每个匹配图像的清晰度进行量化。例如：清晰度的总范围为0％-100％，从0％-100％清晰度递增，0％为不清晰，100％为最清晰；将清晰度划分为5个等级；第1等级清晰度的范围为0％-20％，第2等级清晰度的范围为21％-40％，第3等级清晰度的范围为41％-60％，第4等级清晰度的范围为61％-80％，第5等级清晰度的范围为81％-100％。
步骤S360，分别展示每个图像组中的每个匹配图像及其对应的清晰度。
进一步地，如果已经对清晰度进行了量化，则展示每个图像组中的每个匹配图像及其对应的清晰度等级。
如果一个图像组中包含两组场景相同的图像，则可以每次仅展示一组场景相同的图像及其清晰度(或清晰度等级)。
步骤S370，根据每个匹配图像及其对应的清晰度，选择删除或保留该图像。
例如：清晰度阈值为50％，当图像的清晰度小于50％时，则直接删除图像。又如：在多个匹配图像中，对图像的清晰度进行排序，选择保留清晰度最高的图像、删除其他匹配图像。再如：用户根据展示的匹配图像以及匹配图像的清晰度，手动保留或删除图像。
在一个实施例中，还可以根据图像中图像信号频率的变化，计算图像的噪点个数，将图像的噪点个数和图像一同展示，作为图像保留或删除的参考数据。
针对步骤S330具体而言，
图4是根据本发明一实施例的计算图像之间的匹配度的步骤的流程图。
本实施例引入安卓(Android)版本的OpenCV(Open Source Computer
Vision Library)函数库。通过安卓自带函数以及OpenCV函数库中的函数，来计算图像之间的匹配度。
步骤S410，获取图像组中每个图像的缩略图。
图像的缩略图是指：图像经压缩处理后得到的像素较低的小图。
通过ThumbnailUtils类，获取图像的缩略图。该ThumbnailUtils类为安卓自带函数类。例如：通过以下语句来获取图像的缩略图：
Bitmap bitmap＝ThumbnailUtils.extractThumbnail(bitmap,width,height,ThumbnailUtils.OPTIONS_RECYCLE_INPUT)
步骤S420，在每个图像的缩略图中，提取视觉图像特征点。
视觉图像特征点是指：图像的局部特征。例如：视觉图像特征点为：尺度不变特征转换(Scale-invariant feature transform，SIFT)特征点、或者SURF(Speed up robust feature，SURF)特征点。
以SIFT特征点为例，SIFT特征为图像处理领域的一种描述子，这种描述子具有尺度不变性，可在图像中检测出关键点；SIFT是一个128维的向量(特征向量)。
可以通过OpenCV函数库中的SIFT算法来提取视觉图像特征点，例如：通过以下语句来提取缩略图中的视觉图像特征点：
featuredetector.detect(BitmapMat,keypoints)
步骤S430，基于每个图像的视觉图像特征点，计算图像组中图像两两之间的匹配度。
通过计算两个图像中各视觉图像特征点的欧几里得距离，来获得该两个图像的匹配度。因为，利用SIFT特征来计算两个图像的匹配度为现有技术，在此只举一个例子，使本实施例更加便于理解，具体的技术细节不作赘述。
例如：基于第一图像中的视觉图像特征点k1，在第二图像中找出距离该视觉图像特征点k1距离最近的两个视觉图像特征点(第一关键点和第二关键点)，如果第一关键点与视觉图像特征点k1的距离小于等于第二关键点与视觉图像特征点k1的距离，则用第一关键点对应的距离除以第二关键点对应的距离，如果获得的值大于预设的比例阈值，则说明第一图像和第二图像匹配。
本发明还提供了一种图像处理装置。该图像处理装置可以设置在终端中，终端该图像处理装置，对存储在终端中的图像进行图像处理。
如图5所示，为根据本发明一实施例的图像处理装置的结构图。
在该装置中，包括：
计算模块510，用于计算待处理的图像之间的匹配度。
获取模块520，用于获取匹配度大于预设的匹配度阈值的多个图像。
处理模块530，用于根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
在一个实施例中，计算模块510，用于在待处理的图像中，根据每个图像的属性，将所述待处理的图像划分为一个或多个图像组；分别计算每个图像组中的图像之间的匹配度；其中，所述图像的属性包括：可交换图像文件EXIF属性；其中，所述EXIF属性包括图像拍摄时间信息和图像拍摄位置信息；所述计算模块，根据每个图像的EXIF属性，将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的多个图像，划分为一个图像组。获取模块520，用于分别获取每个图像组中匹配度大于预设的匹配度阈值的多个图像。
具体的，计算模块510，用于针对每个图像组，获取所述图像组中每个图像的缩略图；在每个图像的缩略图中，提取视觉图像特征点；基于每个图像的视觉图像特征点，计算所述图像组中图像两两之间的匹配度。
在另一实施例中，处理模块530，用于针对所述匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；根据每个图像的清晰度和/或噪点个数，选择删除或保留所述多个图像中的全部或部分图像。
本发明还提供了一种终端。如图所示，所述终端可以包括：至少一个输入装置603，至少一个输出装置604，至少一个处理器601，例如CPU，存储器605和至少一个总线602。
其中，上述总线602用于连接上述输入装置603、输出装置604、处理器601和存储器605。
上述存储器605可以是高速RAM存储器，也可为非不稳定的存储器(non-volatile memory)，例如磁盘存储器。上述存储器605还用于存储一组程序代码，上述输入装置603、输出装置604和处理器601用于调用存储器605中存储的程序代码，执行如下操作：
所述处理器601，用于在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度；
所述输入装置603，用于获取匹配度大于预设的匹配度阈值的多个图像；
所述处理器601，还用于根据预设的图像管理策略，通过所述输出装置604删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
在另一实施例中，所述处理器601在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度，包括：
所述处理器601在待处理的图像中，根据每个图像的属性，将所述待处理的图像划分为一个或多个图像组；分别计算每个图像组中的图像之间的匹配度；其中，所述图像的属性包括：可交换图像文件EXIF属性；其中，所述EXIF属性包括图像拍摄时间信息和图像拍摄位置信息；所述处理器601根据每个图像的EXIF属性，将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的多个图像，划分为一个图像组；
所述输入装置603分别获取每个图像组中匹配度大于预设的匹配度阈值的多个图像。
在另一实施例中，所述处理器601分别计算每个图像组中的图像之间的匹配度，包括：
针对每个图像组，所述输入装置603获取所述图像组中每个图像的缩略图；
所述处理器601在每个图像的缩略图中，提取视觉图像特征点；
所述处理器601基于每个图像的视觉图像特征点，计算所述图像组中图像
两两之间的匹配度。
在另一实施例中，所述处理器601根据预设的图像管理策略，通过所述输出装置604删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像，包括：
所述处理器601针对所述匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；
所述处理器601根据每个图像的清晰度和/或噪点个数，通过所述输出装置604选择删除或保留所述多个图像中的全部或部分图像。
具体的，本发明实施例中介绍的终端可以用以实施本发明结合图2～图4介绍的方法实施例中的部分或全部流程。
本发明所述的装置的功能已经在图2-图4所示的方法实施例中进行了描述，故本实施例的描述中未详尽之处，可以参见前述实施例中的相关说明，在此不做赘述。
以上所述仅为本发明的实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的权利要求范围之内。
Claims (13)
- 一种图像处理方法，其特征在于，包括：在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度；获取匹配度大于预设的匹配度阈值的多个图像；根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
- 根据权利要求1所述的方法，其特征在于，在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度，包括：在待处理的图像中，根据每个图像的属性，将所述待处理的图像划分为一个或多个图像组；分别计算每个图像组中的图像之间的匹配度；所述获取匹配度大于预设的匹配度阈值的多个图像，包括：分别获取每个图像组中匹配度大于预设的匹配度阈值的多个图像。
- 根据权利要求2所述的方法，其特征在于，所述图像的属性包括：可交换图像文件EXIF属性；其中，所述EXIF属性包括图像拍摄时间信息和图像拍摄位置信息；根据每个图像的属性，将待处理的多个图像划分为一个或多个图像组，包括：根据每个图像的EXIF属性，将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的多个图像，划分为一个图像组。
- 根据权利要求2所述的方法，其特征在于，所述分别计算每个图像组中的图像之间的匹配度，包括：针对每个图像组，获取所述图像组中每个图像的缩略图；在每个图像的缩略图中，提取视觉图像特征点；基于每个图像的视觉图像特征点，计算所述图像组中图像两两之间的匹配度。
- 根据权利要求1所述的方法，其特征在于，所述根据预设的图像管理 策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像，包括：针对所述匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；根据每个图像的清晰度和/或噪点个数，选择删除或保留所述多个图像中的全部或部分图像。
- 一种图像处理装置，其特征在于，包括：计算模块，用于在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度；获取模块，用于获取匹配度大于预设的匹配度阈值的多个图像；处理模块，用于根据预设的图像管理策略，删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
- 根据权利要求6所述的装置，其特征在于，所述计算模块，用于在待处理的图像中，根据每个图像的属性，将所述待处理的图像划分为一个或多个图像组；分别计算每个图像组中的图像之间的匹配度；其中，所述图像的属性包括：可交换图像文件EXIF属性；其中，所述EXIF属性包括图像拍摄时间信息和图像拍摄位置信息；所述计算模块，根据每个图像的EXIF属性，将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的多个图像，划分为一个图像组；所述获取模块，用于分别获取每个图像组中匹配度大于预设的匹配度阈值的多个图像。
- 根据权利要求7所述的装置，其特征在于，所述计算模块，用于：针对每个图像组，获取所述图像组中每个图像的缩略图；在每个图像的缩略图中，提取视觉图像特征点；基于每个图像的视觉图像特征点，计算所述图像组中图像两两之间的匹配度。
- 根据权利要求6所述的装置，其特征在于，所述处理模块，用于：针对所述匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；根据每个图像的清晰度和/或噪点个数，选择删除或保留所述多个图像中的全部或部分图像。
- 一种终端，其特征在于，所述终端包括通信总线、输入装置、输出装置、存储器以及处理器，其中：所述通信总线，用于实现所述输入装置、输出装置、存储器以及处理器之间的连接通信；所述存储器中存储一组程序代码，且处理器调用存储器中存储的程序代码，用于执行以下操作：所述处理器，用于在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度；所述输入装置，用于获取匹配度大于预设的匹配度阈值的多个图像；所述处理器，还用于根据预设的图像管理策略，通过所述输出装置删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像。
- 根据权利要求11所述的终端，其特征在于，所述处理器在待处理的多个图像中，根据每个图像的特征，计算图像之间的匹配度，包括：所述处理器在待处理的图像中，根据每个图像的属性，将所述待处理的图像划分为一个或多个图像组；分别计算每个图像组中的图像之间的匹配度；其中，所述图像的属性包括：可交换图像文件EXIF属性；其中，所述EXIF属性包括图像拍摄时间信息和图像拍摄位置信息；所述处理器根据每个图像的EXIF属性，将拍摄距离小于预设的距离阈值、且拍摄时间间隔小于预设的时间阈值的多个图像，划分为一个图像组；所述输入装置分别获取每个图像组中匹配度大于预设的匹配度阈值的多个图像。
- 根据权利要求12所述的终端，其特征在于，所述处理器分别计算每个图像组中的图像之间的匹配度，包括：针对每个图像组，所述输入装置获取所述图像组中每个图像的缩略图；所述处理器在每个图像的缩略图中，提取视觉图像特征点；所述处理器基于每个图像的视觉图像特征点，计算所述图像组中图像两两之间的匹配度。
- 根据权利要求11所述的终端，其特征在于，所述处理器根据预设的图像管理策略，通过所述输出装置删除或保留所述匹配度大于预设的匹配度阈值的多个图像中的全部或部分图像，包括：所述处理器针对所述匹配度大于预设的匹配度阈值的多个图像，计算其中每个图像的清晰度和/或噪点个数；所述处理器根据每个图像的清晰度和/或噪点个数，通过所述输出装置选择删除或保留所述多个图像中的全部或部分图像。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510082223.1 | 2015-02-15 | ||
| CN201510082223.1A CN105989599A (zh) | 2015-02-15 | 2015-02-15 | 一种图像处理方法、装置和终端 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016127478A1 true WO2016127478A1 (zh) | 2016-08-18 |
Family
ID=56614233
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2015/075563 WO2016127478A1 (zh) | 2015-02-15 | 2015-03-31 | 一种图像处理方法、装置和终端 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105989599A (zh) |
| WO (1) | WO2016127478A1 (zh) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109064385A (zh) * | 2018-06-20 | 2018-12-21 | 何中 | 360度全景展示效果生成工具及发布系统 |
| US11403336B2 (en) * | 2005-10-26 | 2022-08-02 | Cortica Ltd. | System and method for removing contextually identical multimedia content elements |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106777007A (zh) * | 2016-12-07 | 2017-05-31 | 北京奇虎科技有限公司 | 相册分类优化方法、装置及移动终端 |
| CN106657817A (zh) * | 2016-12-28 | 2017-05-10 | 杭州趣维科技有限公司 | 一种应用于手机平台的自动制作相册mv的处理方法 |
| CN107241504B (zh) * | 2017-06-08 | 2020-03-27 | 努比亚技术有限公司 | 一种图像处理方法、移动终端和计算机可读存储介质 |
| CN107295254B (zh) * | 2017-06-21 | 2021-01-15 | 深圳传音通讯有限公司 | 照片处理方法与拍照终端 |
| CN109993687B (zh) * | 2017-12-29 | 2023-06-13 | 阿里巴巴集团控股有限公司 | 一种图像信息处理方法及装置 |
| CN108182271B (zh) * | 2018-01-18 | 2020-11-17 | 维沃移动通信有限公司 | 一种拍照方法、终端及计算机可读存储介质 |
| CN108413579B (zh) * | 2018-03-07 | 2021-02-05 | 江海军 | 一种乐器房空调器及其控制方法 |
| CN108317693B (zh) * | 2018-03-07 | 2021-01-19 | 范雪梅 | 一种乐器房空调器及其控制方法 |
| JP7024603B2 (ja) * | 2018-05-23 | 2022-02-24 | トヨタ自動車株式会社 | データ記録装置 |
| CN111597363A (zh) * | 2019-02-21 | 2020-08-28 | 奇酷互联网络科技（深圳）有限公司 | 检测重复照片的方法、移动终端和具有存储功能的装置 |
| CN109981903B (zh) * | 2019-03-27 | 2020-12-18 | 联想(北京)有限公司 | 一种图像处理方法及电子设备 |
| CN110704380A (zh) * | 2019-08-27 | 2020-01-17 | 努比亚技术有限公司 | 一种图片自动删除方法、终端及计算机可读存储介质 |
| CN110795579B (zh) * | 2019-10-29 | 2022-11-18 | Oppo广东移动通信有限公司 | 图片清理方法、装置、终端及存储介质 |
| CN111309957A (zh) * | 2020-03-17 | 2020-06-19 | 杭州趣维科技有限公司 | 一种自动生成旅行相册mv的方法 |
| CN111669495B (zh) * | 2020-05-29 | 2021-11-12 | 维沃移动通信有限公司 | 拍照方法、拍照装置和电子设备 |
| CN112749131A (zh) * | 2020-06-11 | 2021-05-04 | 腾讯科技（上海）有限公司 | 信息消重处理方法、装置及计算机可读存储介质 |
| CN111623783A (zh) * | 2020-06-30 | 2020-09-04 | 杭州海康机器人技术有限公司 | 一种初始定位方法、视觉导航设备、仓储系统 |
| WO2022032564A1 (zh) * | 2020-08-13 | 2022-02-17 | 华为技术有限公司 | 白平衡处理的方法与装置 |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101290634A (zh) * | 2008-06-03 | 2008-10-22 | 北京搜狗科技发展有限公司 | 一种识别重复缩略图的方法、装置及在搜索引擎中的应用 |
| CN102509285A (zh) * | 2011-09-28 | 2012-06-20 | 宇龙计算机通信科技(深圳)有限公司 | 一种拍摄出模糊图片的处理方法、系统及拍摄设备 |
| CN103412753A (zh) * | 2013-08-12 | 2013-11-27 | 宇龙计算机通信科技(深圳)有限公司 | 图片删除方法及装置 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1909672A (zh) * | 2006-08-18 | 2007-02-07 | 北京金山软件有限公司 | 一种用于显示图片的存储方法及系统 |
| JP6264965B2 (ja) * | 2014-03-14 | 2018-01-24 | オムロン株式会社 | 画像処理装置、画像処理方法、画像処理プログラム |
-
2015
- 2015-02-15 CN CN201510082223.1A patent/CN105989599A/zh active Pending
- 2015-03-31 WO PCT/CN2015/075563 patent/WO2016127478A1/zh active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101290634A (zh) * | 2008-06-03 | 2008-10-22 | 北京搜狗科技发展有限公司 | 一种识别重复缩略图的方法、装置及在搜索引擎中的应用 |
| CN102509285A (zh) * | 2011-09-28 | 2012-06-20 | 宇龙计算机通信科技(深圳)有限公司 | 一种拍摄出模糊图片的处理方法、系统及拍摄设备 |
| CN103412753A (zh) * | 2013-08-12 | 2013-11-27 | 宇龙计算机通信科技(深圳)有限公司 | 图片删除方法及装置 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11403336B2 (en) * | 2005-10-26 | 2022-08-02 | Cortica Ltd. | System and method for removing contextually identical multimedia content elements |
| CN109064385A (zh) * | 2018-06-20 | 2018-12-21 | 何中 | 360度全景展示效果生成工具及发布系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105989599A (zh) | 2016-10-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016127478A1 (zh) | 一种图像处理方法、装置和终端 | |
| US10284789B2 (en) | Dynamic generation of image of a scene based on removal of undesired object present in the scene | |
| US9665962B2 (en) | Image distractor detection and processng | |
| JP5612310B2 (ja) | 顔認識のためのユーザーインターフェース | |
| CN106899781B (zh) | 一种图像处理方法及电子设备 | |
| CN108241645B (zh) | 图像处理方法及装置 | |
| US10277806B2 (en) | Automatic image composition | |
| US9426385B2 (en) | Image processing based on scene recognition | |
| JP2016521892A5 (zh) | ||
| US20160323505A1 (en) | Photographing processing method, device and computer storage medium | |
| US10455163B2 (en) | Image processing apparatus that generates a combined image, control method, and storage medium | |
| WO2022160857A1 (zh) | 图像处理方法及装置、计算机可读存储介质和电子设备 | |
| CN114283156B (zh) | 一种用于去除文档图像颜色及手写笔迹的方法及装置 | |
| US20170019615A1 (en) | Image processing method, non-transitory computer-readable storage medium and electrical device thereof | |
| US9767533B2 (en) | Image resolution enhancement based on data from related images | |
| US20170083545A1 (en) | Image extraction system, image extraction method, image extraction program, and recording medium storing program | |
| CN110727810B (zh) | 图像处理方法、装置、电子设备及存储介质 | |
| JP2008035149A (ja) | 映像記録再生システムおよび映像記録再生方法 | |
| WO2015196681A1 (zh) | 一种图片处理方法及电子设备 | |
| US9094617B2 (en) | Methods and systems for real-time image-capture feedback | |
| KR20160059358A (ko) | 영상 처리 방법, 상기 방법을 기록한 컴퓨터 판독 가능 저장매체 및 전자 기기. | |
| US10373329B2 (en) | Information processing apparatus, information processing method and storage medium for determining an image to be subjected to a character recognition processing | |
| US20190188480A1 (en) | Automated video correction | |
| US9489727B2 (en) | Method for generating a preferred image by replacing a region of a base image | |
| CN113766123B (zh) | 一种拍照美化方法及终端 |
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: 15881639 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 10/01/2018) |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 15881639 Country of ref document: EP Kind code of ref document: A1 |