WO2018223318A1 - 360度全景图像及360度全景视频的识别方法及电子装置 - Google Patents

360度全景图像及360度全景视频的识别方法及电子装置 Download PDF

Info

Publication number
WO2018223318A1
WO2018223318A1 PCT/CN2017/087475 CN2017087475W WO2018223318A1 WO 2018223318 A1 WO2018223318 A1 WO 2018223318A1 CN 2017087475 W CN2017087475 W CN 2017087475W WO 2018223318 A1 WO2018223318 A1 WO 2018223318A1
Authority
WO
WIPO (PCT)
Prior art keywords
pixel
pixels
gray value
column
difference
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2017/087475
Other languages
English (en)
French (fr)
Inventor
卢启栋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Royole Technologies Co Ltd
Original Assignee
Shenzhen Royole Technologies Co Ltd
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 Shenzhen Royole Technologies Co Ltd filed Critical Shenzhen Royole Technologies Co Ltd
Priority to PCT/CN2017/087475 priority Critical patent/WO2018223318A1/zh
Priority to CN201780004657.7A priority patent/CN108475432B/zh
Publication of WO2018223318A1 publication Critical patent/WO2018223318A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/30Determination of transform parameters for the alignment of images, i.e. image registration

Definitions

  • the invention relates to a video image recognition method, in particular to a 360-degree panoramic image recognition method and a 360-degree panoramic video recognition method, and an electronic device applying the same.
  • 360-degree panoramic camera technology has been widely used, and 360-degree panoramic image or 360-degree panoramic video can be obtained through 360-degree panoramic photography technology.
  • 360-degree panoramic image or 360-degree panoramic video can be obtained through 360-degree panoramic photography technology.
  • the existing recognition method is usually based on the aspect ratio of the image frame in the image file or the video file and/or the RGB (red, green and blue primary colors) values for judgment and recognition, which is often not accurate enough.
  • the embodiment of the invention discloses a 360-degree panoramic image and a 360-degree panoramic video recognition method and an electronic device, which can accurately recognize a 360-degree panoramic image or a 360-degree panoramic video.
  • the method for identifying a 360-degree panoramic image disclosed by the embodiment of the present invention includes: acquiring an image file to be identified; determining a target image frame according to the image file to be identified; and determining pixels of the first column of the target image frame Whether the degree of matching between the gray value of the point and the gray value of the pixel of the last column is greater than or equal to the first threshold, and determining the difference between the gray values of all the pixels of the first line of the target image frame and Whether the difference between the gray values of all the pixels of the last row is less than or equal to the second threshold; if yes, determining that the image file to be identified is a 360-degree panoramic image; otherwise, determining that the image file to be identified is not 360 degree panoramic image.
  • the method for identifying a 360-degree panoramic video disclosed in the embodiment of the present invention includes: acquiring a video file to be identified; extracting a plurality of representative image frames from the video file to be identified; determining whether each representative image frame is a 360-degree panoramic image; whether the proportion of the representative image frame determined as the 360-degree panoramic image is greater than a preset ratio; if yes, determining that the video file is a 360-degree panoramic video, and if not, determining that the video file is not 360 Degree panoramic video.
  • the electronic device disclosed in the embodiment of the present invention includes a processor and a memory, wherein the memory stores a video file to be identified, the processor is configured to: acquire a video file to be identified; and extract a number of the video files to be identified Representing an image frame; determining whether each representative image frame is a 360-degree panoramic image; determining whether a proportion of the representative image frame of the 360-degree panoramic image is greater than a preset ratio; if yes, determining that the video file is a 360-degree panoramic video If not, it is determined that the video file is not a 360-degree panoramic video.
  • the electronic device disclosed in the embodiment of the present invention includes a processor and a memory, wherein the memory stores an image file to be identified, the processor is configured to: acquire an image file to be identified; and determine a target image frame according to the image file to be identified; Determining whether a matching degree between a gray value of a pixel of the first column of the target image frame and a gray value of a pixel of the last column is greater than or equal to a first threshold, and determining a first of the target image frame Whether the difference between the gray value of all the pixels of the row and the difference of the gray values of all the pixels of the last row are less than or equal to the second threshold; if yes, determining that the image file to be recognized is a 360-degree panoramic image, Otherwise, it is determined that the image file to be identified is not a 360-degree panoramic image.
  • the 360-degree panoramic image and the 360-degree panoramic video identification method and the electronic device of the present invention determine the target image frame, and then analyze the gray-scale difference between the pixel of the first column of the target image frame and the pixel of the last column. And determining a grayscale difference between pixels of the first row of the target image frame and a grayscale difference between the pixels of the last row to determine whether the target image frame is a 360-degree panoramic image frame, and then determining the corresponding image Whether the video is a 360-degree panoramic image or a 360-degree panoramic video can effectively improve the recognition accuracy.
  • FIG. 1 is a flowchart of a method for recognizing a 360-degree panoramic image according to an embodiment of the present invention.
  • FIG. 2 is a schematic diagram of a target image frame in an embodiment of the present invention.
  • FIG. 3 is a flowchart of a method for recognizing a 360-degree panoramic image according to another embodiment of the present invention.
  • step S12 of FIG. 1 or step S102 of FIG. 3 in an embodiment of the present invention.
  • FIG. 5 is a sub-flow diagram of step S104 of FIG. 3 in an embodiment of the present invention.
  • FIG. 6 is a flowchart of a method for recognizing a 360-degree panoramic video according to an embodiment of the present invention.
  • Figure 7 is a sub-flow diagram of an embodiment in step S503 of Figure 6.
  • FIG. 8 is a structural block diagram of an electronic device according to an embodiment of the present invention.
  • FIG. 1 is a flowchart of a method for recognizing a 360-degree panoramic image according to an embodiment of the present invention.
  • 2 is a schematic diagram of a target image frame M1. The identification method includes the steps of:
  • S12 Determine the target image frame M1 according to the image file to be identified.
  • S13 determining whether the matching degree between the gray value of the pixel P1 of the first column L1 of the target image frame M1 and the gray value of the pixel of the last column Ln is greater than or equal to a first threshold, and determining the location Whether the difference between the gray value of all the pixel points P1 of the first line C1 of the target image frame M1 and the difference of the gray values of all the pixel points P1 of the last line C1 is less than or equal to the second threshold.
  • step S14 is performed, and if not, the pixel of the first column L1
  • the degree of matching between the gray value of the point P1 and the gray value of the pixel of the last column Ln is smaller than the first threshold, or The difference between the gray value of all the pixel points P1 of the first row C1 and the gray value of all the pixel points P1 of the last row C1 are greater than the second threshold, and then step S15 is performed.
  • the present application it is determined by determining that the matching degree between the gradation value of the pixel point P1 of the first column L1 and the gradation value of the pixel point P1 of the last column L1 in the target image frame M1 is greater than the first threshold. Whether the gradation value of the pixel point P1 of the first column L1 of the target image frame is similar to the gradation value of the pixel point P1 of the last column Ln.
  • FIG. 3 is a flowchart of a method for recognizing a 360-degree panoramic image according to another embodiment of the present invention.
  • the method for identifying the 360-degree panoramic image is not limited to the following execution sequence.
  • the identifying method comprises:
  • S101 Acquire an image file to be identified.
  • the target image frame M1 includes pixel points P1 arranged in a matrix.
  • the target image frame M1 includes m*n pixel points P1, that is, m rows and n columns of pixel points P1, where m and n are positive integers greater than 1, and m and n may be equal or not equal.
  • S104 Calculate a matching degree between the gradation value of the pixel point P1 of the first column L1 and the gradation value of the pixel point of the last column Ln.
  • step S105 Determine whether the matching degree is greater than or equal to a first threshold. If yes, step S106 is performed, and if no, step S110 is performed.
  • S106 Acquire gray values of all the pixel points P1 of the first row C1 of the target image frame M1 and gray values of all the pixel points P1 of the last row of Cm.
  • the step S107 Calculate a difference between the gradation values of all the pixel points P1 of the first row C1 and the gradation values of all the pixel points P1 of the last row Cm.
  • the step S107 Specifically, the method includes: calculating a variance of the gray values of all the pixel points P1 of the first row C1 to obtain a difference between the gray values of all the pixel points P1, and calculating a gray value of all the pixel points P1 of the last row Cm. The variance obtains the difference between the gray values of all the pixel points P1 of the last line Cm.
  • step S108 Determine whether the difference between the gray value of all the pixel points P1 of the first row C1 and the difference of the gray values of all the pixel points P1 of the last row C1 are less than or equal to the second threshold. If yes, step S109 is performed, and if no, step S110 is performed.
  • the second threshold may be a variance threshold of 20.
  • S109 Determine that the image file to be identified is a 360-degree panoramic image.
  • steps S103-S108 correspond to the step S13 in FIG. 1, and the more specific steps of step S13 are illustrated. It should be noted that when steps S103-S108 are taken as a more specific step flow of step S13, step S106 will be executed regardless of whether the determination result of S105 is YES or not.
  • step S12 of FIG. 1 or step S102 of FIG. 3 in an embodiment of the present invention.
  • the step S12 or the step S102 includes:
  • step S1021 It is judged whether the image to be recognized is a three-dimensional image or a two-dimensional image (step S1021). If it is a three-dimensional image, step S1023 is performed, and if it is a two-dimensional image, step S1025 is performed.
  • the left-eye image or the right-eye image in the image to be recognized is extracted, and it is determined that the extracted left-eye image or right-eye image is the target image frame M1 (S1023).
  • the image to be recognized is the target image frame M1 (S1025).
  • the present application determines whether the three-dimensional image is a three-dimensional image, and extracts a left-eye image or a right image when determining the three-dimensional image.
  • the eye image is analyzed as a target image frame to improve the accuracy of the analysis result.
  • step S104 includes:
  • Step S1041 sequentially, each pixel point P1 of the first column L1 and/or the last column Ln is taken as a center point, and the pixel point P1 of the first column L1 and/or the last column Ln which is the current center point is obtained.
  • Step S1043 Calculate an average value of the obtained grayscale values of the plurality of pixel points P1 including the pixel point P1 that is currently the center point, and use the average value as the pixel point P1 of the current center point.
  • the gradation value is corrected, and the corrected gradation value of each pixel point P1 of the first column L1 and the last column Ln is obtained one by one.
  • the gray value of the preset pixel point P1 adjacent to the pixel point P1 of the first column L1 that is currently the center point and the pixel point P1 of the current center point may be acquired first.
  • the pixel value is then calculated as the corrected gray value of the pixel point P1 which is currently the center point, and the corrected gray value of all the pixel points P1 of the first column L1 is obtained, and then the pixel point of the last column Ln which is the current center point is obtained.
  • the gray value of the preset pixel point P1 adjacent to the pixel point P1 of the first column L1 and the last column of the current column and the pixel point of the current center point may be simultaneously acquired. Pixel value of P1, and then simultaneously calculate the corrected gradation value of the pixel point P1 of the first column L1 and the last column of the current center point, and obtain each pixel point P1 of the first column L1 and the last column Ln Correct the gray value.
  • Step S1045 respectively calculating a difference between the corrected gradation value of each pixel point P1 of the first column L1 and the corrected gradation value of the pixel point P1 of the row corresponding to the last column Ln to obtain a plurality of gradation value difference values.
  • Step S1047 Calculate a proportion of the calculated gray value difference values that are less than or equal to the preset difference value among the calculated gray value difference values, and determine the ratio as the matching degree. For example, if the m is 1000, that is, the first column L1 and the last column Lm each include 1000 pixel points P1, 1000 gray value difference values will be calculated, assuming a gray value less than or equal to the preset difference value. The number of differences is 800, and the proportion is 0.8, which is 80%.
  • step S105 in FIG. 1 determines whether the matching degree is greater than or equal to a first threshold value in a first threshold value, for example, 0.75, that is, 75. %.
  • FIG. 6 is a flowchart of a method for recognizing a 360-degree panoramic video according to an embodiment of the present invention. As shown in FIG. 6, the method for identifying the 360-degree panoramic video includes the following steps:
  • S501 Acquire a video file to be identified.
  • S502 Extract a plurality of representative image frames from the video file to be identified.
  • the representative image frame is selected based on at least one of the following selection principles: 1. Selecting a key frame in the video file as a representative image frame, and generally the key frame includes a complete frame image, other than Type frames have higher image quality. Key frames often include the start time of a camera lens. There is a big difference between adjacent key frames, which is beneficial to improve the diversity of recognition samples. 2. Select frames with rich picture changes. The image is used as a representative image frame; 3.
  • a plurality of image frames whose time periods covered by the position of the time axis in the video file exceed a preset value are selected as a plurality of representative image frames, for example, the duration of the video file is 2 hours. Then, as many as possible, a plurality of image frames are intermittently selected as representative image frames from the beginning to the last time period, and the time span of the plurality of image frames may be 1 hour and 50 minutes or the like.
  • step S504 Determine whether a proportion of the representative image frame of the 360-degree panoramic image is greater than a preset ratio. If yes, step S505 is performed, and if no, step S506 is performed.
  • S505 Determine that the video file is a 360-degree panoramic video.
  • S506 Determine that the video file is not a 360-degree panoramic video.
  • step S503 includes:
  • S531 Determine a target image frame corresponding to each representative image frame
  • S532 Determine whether the matching degree between the gray value of the pixel of the first column of each target image frame and the gray value of the pixel of the last column is greater than or equal to a first threshold, and determine each target image frame. Whether the difference between the gray values of all the pixels of the first row and the gray value of all the pixels of the last row are less than or equal to the second threshold. If yes, go to step S533, if no, go to step S534.
  • S533 Determine that the corresponding representative image frame is a 360-degree panoramic image.
  • S534 Determine that the corresponding representative image frame is not a 360-degree panoramic image.
  • each representative image frame is a 360-degree panoramic image.
  • steps S531-S534 correspond to steps S12-S15 in FIG. 1, respectively.
  • the more specific steps of steps S531-S534 are the same as steps S12-S15 in FIG. 1, that is, any one of the representative image frames is equivalent to one image file, and any method for identifying whether the image frame is a 360-degree panoramic image in FIG. versus The method for identifying whether the image file is a 360-degree panoramic image is the same in FIG. 1, and further includes the corresponding method steps in FIGS.
  • step S531 includes determining whether the representative image frame is a three-dimensional image or a two-dimensional image.
  • the step S532 may specifically include the steps S103-S108 shown in FIG. 3 or further include the steps S1041-S1047 shown in FIG. 5.
  • FIG. 8 is a structural block diagram of an electronic device 100 according to an embodiment of the invention.
  • the electronic device 100 includes a processor 10 and a memory 20 .
  • the memory 20 stores image files and/or video files to be identified.
  • the image files and/or video files stored in the memory 20 are pre-stored in the memory 20, and may also be temporarily downloaded from the server or received from other electronic devices 100.
  • the electronic device 100 further includes a communication unit 30, which can establish a communication connection with a server or other electronic device through the communication unit 30 in advance, and receive an image to be recognized from a server or other electronic device.
  • File and/or video files are stored in the memory 20.
  • the communication unit 30 can be a wired or wireless communication module, and can be, for example, a wired network interface unit, a WIFI module, a Bluetooth module, or the like.
  • the processor 10 is configured to analyze an image file and/or a video file to be identified stored in the memory 20 to identify whether it is a 360-degree panoramic image and/or a 360-degree panoramic video.
  • the processor 10 is configured to perform any of the methods shown in FIG. 1 and FIG. 3-5 to identify whether the image file to be identified is a 360-degree panoramic image.
  • the processor 10 is further configured to perform any of the method steps shown in FIG. 6-7 and the related method steps shown in FIG. 3-5 to identify whether the video file to be identified is a 360-degree panoramic video.
  • the memory 20 stores a plurality of program instructions, and after the processor 10 calls the execution of the plurality of program instructions, performing any of the methods shown in FIG. 1 and FIG. 3-5 to identify the to-be-identified Whether the image file is a 360-degree panoramic image, and/or performing any of the method steps shown in FIG. 6-7 and the related method steps shown in FIG. 3-5 to identify whether the video file to be recognized is a 360-degree panoramic video. .
  • the processor 10 performs the following method to identify whether the image file to be recognized is 360 degrees.
  • Panoramic image acquiring an image file to be identified; determining a target image frame according to the image file to be identified; determining a match between a gray value of a pixel of the first column of the target image frame and a gray value of a pixel of the last column Whether the degree is greater than or equal to a first threshold, and whether the difference between the gray values of all the pixels of the first line of the target image frame and the difference of the gray values of all the pixels of the last line are less than or equal to the second threshold If yes, it is determined that the image file to be identified is a 360-degree panoramic image; if not, it is determined that the image file to be recognized is not a 360-degree panoramic image.
  • the processor 10 performs the following method: determining whether the video file to be identified is a 360-degree panoramic video: acquiring a video file to be identified; extracting a plurality of representative image frames from the video file to be identified; determining each representative Whether the image frame is a 360-degree panoramic image; whether the proportion of the representative image frame determined as the 360-degree panoramic image is greater than a preset ratio; if yes, determining that the video file is a 360-degree panoramic video, and if not, determining the video The file is not a 360 degree panoramic video.
  • the processor 40 can be a microcontroller, a microprocessor, a single chip, a digital signal processor, or the like.
  • the memory 20 can be any storage device that can store information such as a memory card, a solid state memory, a micro hard disk, an optical disk, or the like.
  • the present invention further provides a computer readable storage medium having stored therein a plurality of program instructions, which are executed by the processor 10 for execution, and are executed in FIG. Any of the method steps 3-7, thereby identifying whether the image file is a 360-degree panoramic image and/or identifying whether the video file is a 360-degree panoramic video.
  • the computer storage medium is the memory 20, and may be any storage device that can store information such as a memory card, a solid state memory, a micro hard disk, an optical disk, or the like.
  • the electronic device 100 can be a mobile phone, a tablet computer, a notebook computer, a desktop computer, etc., and can also be a head-mounted device such as a smart helmet or smart glasses.
  • the 360 degree panoramic image and the 360 degree panoramic video recognition method and the electronic device of the present invention by determining the target image frame, and then analyzing the gray between the pixel of the first column of the target image frame and the pixel of the last column Determining whether the target image frame is a 360-degree panoramic image frame, and then determining whether the target image frame is a 360-degree panoramic image frame by analyzing the grayscale difference between the pixel points of the first row of the target image frame and the grayscale difference between the pixels of the last row. Whether the corresponding image or video is a 360-degree panoramic image or a 360-degree panoramic video can effectively improve the recognition accuracy.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

本申请公开一种360度全景视频的识别方法,包括:获取待识别的视频文件;从所述待识别的视频文件中提取若干代表图像帧;确定每一代表图像帧是否为360度全景图像;确定为360度全景图像的代表图像帧所占的比例是否大于预设比例;如果是,确定所述视频文件为360度全景视频,如果否,确定所述视频文件不为360度全景视频。本申请还公开一种360度全景图像的识别方法及电子装置。本申请的电子装置及360度全景视频、图像的识别方法,能够以更简单的方式更准确地识别出360度全景视频或360度全景图像。

Description

360度全景图像及360度全景视频的识别方法及电子装置 技术领域
本发明涉及一种视频图像识别方法,尤其涉及一种360度全景图像的识别方法及360度全景视频的识别方法,以及应用所述识别方法的电子装置。
背景技术
目前,360度全景拍照技术已经较为广泛应用,通过360度全景拍照技术可以得到360度全景图像或360度全景视频。在一些情境下,当播放某一图像或视频文件时,通常需要识别是否为360度全景视频,以进行相应的播放设置,体现360度全景的效果。现有的识别方式通常为根据图像文件或视频文件中的图像帧的长宽比和/或RGB(红绿蓝三原色)值来进行判断识别,往往不够准确。
发明内容
本发明实施例公开一种360度全景图像及360度全景视频的识别方法及电子装置,能够准确地识别出360度全景图像或360度全景视频。
本发明实施例公开的360度全景图像的识别方法,所述识别方法包括:获取待识别的图像文件;根据待识别的图像文件确定目标图像帧;判断所述目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断所述目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值;如果是,确定所述待识别的图像文件为360度全景图像,否则,确定所述待识别的图像文件不为360度全景图像。
本发明实施例公开的360度全景视频的识别方法,所述识别方法包括:获取待识别的视频文件;从所述待识别的视频文件中提取若干代表图像帧;确定每一代表图像帧是否为360度全景图像;确定为360度全景图像的代表图像帧所占的比例是否大于预设比例;如果是,确定所述视频文件为360度全景视频,如果否,确定所述视频文件不为360度全景视频。
本发明实施例公开的电子装置,包括处理器以及存储器,所述存储器中存储有待识别的视频文件,所述处理器用于:获取待识别的视频文件;从所述待识别的视频文件中提取若干代表图像帧;确定每一代表图像帧是否为360度全景图像;确定为360度全景图像的代表图像帧所占的比例是否大于预设比例;如果是,确定所述视频文件为360度全景视频,如果否,确定所述视频文件不为360度全景视频。
本发明实施例公开的电子装置,包括处理器以及存储器,所述存储器中存储有待识别的图像文件,所述处理器用于:获取待识别的图像文件;根据待识别的图像文件确定目标图像帧;判断所述目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断所述目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值;如果是,确定所述待识别的图像文件为360度全景图像,否则,确定所述待识别的图像文件不为360度全景图像。
本发明的360度全景图像及360度全景视频的识别方法及电子装置,通过确定目标图像帧,然后分析目标图像帧的第一列的像素点与最后一列的像素点之间的灰度差值以及分析目标图像帧的第一行的像素点之间的灰度差值以及最后一行的像素点之间的灰度差值来判断目标图像帧是否为360度全景图像帧,继而判断对应的图像或视频是否为360度全景图像或360度全景视频,可有效提高识别准确度。
附图说明
为了更清楚地说明本发明实施例中的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为本发明一实施例中的360度全景图像的识别方法的流程图。
图2为本发明一实施例中目标图像帧的示意图。
图3为本发明另一实施例中的360度全景图像的识别方法的流程图。
图4为图1中步骤S12或图3中步骤S102在本发明一实施例中的子流程图。
图5为图3中步骤S104在本发明一实施例中的子流程图。
图6为本发明一实施例中的360度全景视频的识别方法的流程图。
图7为图6中步骤S503在一实施例中的子流程图。
图8为本发明一实施例中的电子装置的结构框图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
请参阅图1及图2,图1为本发明一实施例中的360度全景图像的识别方法的流程图。图2为目标图像帧M1的示意图。所述识别方法包括步骤:
S11:获取待识别的图像文件。
S12:根据待识别的图像文件确定目标图像帧M1。
S13:判断所述目标图像帧M1的第一列L1的像素点P1的灰度值与最后一列Ln的像素点的灰度值之间的匹配度是否大于或等于一第一阈值,以及判断所述目标图像帧M1的第一行C1的所有像素点P1的灰度值之差以及最后一行Cm的所有像素点P1的灰度值之差是否均小于或等于第二阈值。
如果是,即第一列L1的像素点P1的灰度值与最后一列Ln的像素点的灰度值之间的匹配度大于或等于所述第一阈值且所述第一行C1的所有像素点P1的灰度值之差以及所述最后一行Cm的所有像素点P1的灰度值之差均小于或等于所述第二阈值,则执行步骤S14,如果否,即第一列L1的像素点P1的灰度值与最后一列Ln的像素点的灰度值之间的匹配度小于所述第一阈值,或者 所述第一行C1的所有像素点P1的灰度值之差以及所述最后一行Cm的所有像素点P1的灰度值之差均大于第二阈值,则执行步骤S15。
S14:确定所述待识别的图像文件为360度全景图像。
S15:确定所述待识别的图像文件不为360度全景图像。
从而,本申请中,通过判断目标图像帧M1中的第一列L1的像素点P1的灰度值以及最后一列Ln的像素点P1的灰度值之间的匹配度大于第一阈值,来确定目标图像帧第一列L1的像素点P1的灰度值是否与最后一列Ln的像素点P1的灰度值相近似。以及判断目标图像帧M1中第一行C1的像素点P1的灰度值之差以及最后一行Cm的像素点P1的灰度值之差是否在第二阈值内,来确定目标图像帧M1第一行C1的像素点P1的灰度值是否相近似,以及最后一行Cm的像素点P1的灰度值是否相近似,可准确地确定目标图像帧M1是否为360度全景图像继而确定待识别的图像文件是否为360度全景图像。
请参阅图3,图3为本发明另一实施例中的360度全景图像的识别方法的流程图。其中,所述360度全景图像的识别方法并不限于如下的执行顺序。在另一实施例中,所述识别方法包括:
S101:获取待识别的图像文件。
S102:根据待识别的图像文件确定目标图像帧M1。
S103:获取所述目标图像帧M1的第一列L1的像素点P1的灰度值以及最后一列的像素点的灰度值。其中,所述目标图像帧M1包括矩阵式排列的像素点P1。例如,如图2所示,所述目标图像帧M1包括m*n个像素点P1,即m行n列的像素点P1,其中m、n为大于1的正整数,m和n可相等或不相等。
S104:计算第一列L1的像素点P1的灰度值与最后一列Ln的像素点的灰度值之间的匹配度。
S105:判断所述匹配度是否大于或等于一第一阈值。如果是,则执行步骤S106,如果否,则执行步骤S110。
S106:获取所述目标图像帧M1的第一行C1的所有像素点P1的灰度值以及最后一行Cm的所有像素点P1的灰度值。
S107:计算所述第一行C1的所有像素点P1的灰度值之差以及所述最后一行Cm的所有像素点P1的灰度值之差。在一些实施例中,所述步骤S107 具体包括:计算所述第一行C1的所有像素点P1的灰度值的方差得到所有像素点P1的灰度值之差,以及计算所述最后一行Cm的所有像素点P1的灰度值的方差得到所述最后一行Cm的所有像素点P1的灰度值之差。
S108:判断所述第一行C1的所有像素点P1的灰度值之差以及所述最后一行Cm的所有像素点P1的灰度值之差是否均小于或等于第二阈值。如果是,则执行步骤S109,如果否,则执行步骤S110。其中,所述第二阈值可为方差阈值20。
S109:确定所述待识别的图像文件为360度全景图像。
S110:确定所述待识别的图像文件不为360度全景图像。
其中,所述步骤S103-S108对应为图1中的步骤S13,示意出了步骤S13更具体的步骤。需要注意的是,当步骤S103-S108作为步骤S13更具体的步骤流程时,不论S105的判断结果为是还是否,都将执行步骤S106。
请一并参阅图4,为图1中步骤S12或图3中步骤S102在本发明一实施例中的子流程图。在一实施例中,所述步骤S12或步骤S102包括:
判断所述待识别的图像为三维图像还是二维图像(步骤S1021)。如果为三维图像,则执行步骤S1023,如果为二维图像,则执行步骤S1025。
如果为三维图像,提取所述待识别的图像中的左眼图像或右眼图像,确定所述提取出来的左眼图像或右眼图像为目标图像帧M1(S1023)。
如果为二维图像,确定所述待识别的图像为所述目标图像帧M1(S1025)。
由于三维图像包括左眼图像和右眼图像,直接对三维图像进行分析会造成结果的不准确,本申请通过预先判断是否为三维图像,在判断为三维图像时,从中提取出左眼图像或右眼图像作为目标图像帧进行分析,可提高分析结果的准确性。
请一并参阅图5,为步骤S104在本发明一实施例中的子流程图。在一实施例中,所述步骤S104包括:
步骤S1041:依次以第一列L1和/或最后一列Ln的每一个像素点P1为中心点,获取所述第一列L1和/或最后一列Ln的当前为中心点的像素点P1上下相邻的预设个像素点P1的灰度值以及所述当前为中心点的像素点P1的像素值。
步骤S1043:计算所述获取到的包括当前为中心点的像素点P1在内的若干像素点P1的灰度值的平均值,将所述平均值作为所述当前为中心点的像素点P1的校正灰度值,以此一一得出所述第一列L1和最后一列Ln的每一个像素点P1的校正灰度值。
其中,在一些实施例中,可以先获取第一列L1的当前为中心点的像素点P1上下相邻的预设个像素点P1的灰度值以及所述当前为中心点的像素点P1的像素值,然后计算当前为中心点的像素点P1的校正灰度值,得出第一列L1的所有像素点P1的校正灰度值,然后再获取最后一列Ln的当前为中心点的像素点P1上下相邻的预设个像素点P1的灰度值以及所述当前为中心点的像素点P1的像素值,然后计算当前为中心点的像素点P1的校正灰度值,得出最后一列Ln的所有像素点P1的校正灰度值,而得出所述第一列L1和最后一列Ln的每一个像素点P1的校正灰度值。
在另一些实施例中,可以同时获取第一列L1及最后一列的当前为中心点的像素点P1上下相邻的预设个像素点P1的灰度值以及所述当前为中心点的像素点P1的像素值,然后同时计算第一列L1及最后一列的当前为中心点的像素点P1的校正灰度值,而得出所述第一列L1和最后一列Ln的每一个像素点P1的校正灰度值。
步骤S1045:分别计算第一列L1的每一个像素点P1的校正灰度值与最后一列Ln对应行的像素点P1的校正灰度值的差值而得出若干个灰度值差值。
步骤S1047:计算所述计算出的若干个灰度值差值中小于或等于预设差值的灰度值差值所占的比例,将所述比例确定为所述匹配度。例如,设所述m为1000,即第一列L1及最后一列Lm均包括1000个像素点P1,则将计算出1000个灰度值差值,假设小于或等于预设差值的灰度值差值的个数为800,则所占的比例为0.8,即80%。
与所述图5相对应,在一些实施例中,图1中所述步骤S105确定所述匹配度是否大于或等于一第一阈值中的第一阈值为一比例值,例如为0.75,即75%。
请参阅图6,为本发明一实施例中的360度全景视频的识别方法的流程图。如图6所示,所述360度全景视频的识别方法包括步骤:
S501:获取待识别的视频文件。
S502:从所述待识别的视频文件中提取若干代表图像帧。在一些实施例中,所述代表图像帧至少基于以下选取原则中的至少一种进行选取:1、选取视频文件中的关键帧作为代表图像帧,一般关键帧包含一个完整的帧画面,比其他类型帧具有更高的图像质量,关键帧往往包括一个摄像镜头的开始时段,相邻的关键帧之间有较大的差异,有利于提高识别样本的多样性;2、选取画面变化丰富的帧图像作为代表图像帧;3、选取多个位于视频文件中的时间轴的位置覆盖的时间段超过预设值的多个图像帧作为多个代表图像帧,例如,设视频文件时长为2小时,则尽可能从开始到最后的时间段内间隔性地选择多个图像帧作为代表图像帧,所述多个图像帧的时间跨度可以为1小时50分钟等。
S503:确定每一代表图像帧是否为360度全景图像。
S504:确定为360度全景图像的代表图像帧所占的比例是否大于预设比例。如果是,则执行步骤S505,如果否,则执行步骤S506。
S505:确定所述视频文件为360度全景视频。
S506:确定所述视频文件不为360度全景视频。
请参阅图7,为步骤S503在一实施例中的子流程图。在一些实施例中,所述步骤S503包括:
S531:确定每一代表图像帧对应的目标图像帧;
S532:判断每一目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于一第一阈值,以及判断每一目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值。如果是,执行步骤S533,如果否,执行步骤S534。
S533:确定对应的代表图像帧为360度全景图像。
S534:确定所述对应的代表图像帧不为360度全景图像。
从而,一一确定每一代表图像帧是否为360度全景图像。
其中,所述步骤S531-S534分别对应图1中的步骤S12-S15。步骤S531-S534的更具体的步骤与图1中的步骤S12-S15相同,即,任一个代表图像帧相当于一个图像文件,图7中识别任一个代表图像帧是否为360度全景图像的方法与 图1中识别所述图像文件是否为360度全景图像的方法相同,同时也进一步包括图3-图5中的相应方法步骤。例如,步骤S531包括:判断所述代表图像帧为三维图像还是二维图像。如果为三维图像,提取所述代表图像帧中的左眼图像或右眼图像,确定所述提取出来的左眼图像或右眼图像为目标图像帧;如果所述代表图像帧为二维图像,确定所述代表图像帧为所述目标图像帧。
相应的,所述步骤S532可具体包括图3所示的步骤S103-S108,或者进一步包括图5中所示的步骤S1041-S1047。
请参阅图8,为本发明一实施例中的电子装置100的结构框图。如图1所示,所述电子装置100包括处理器10及存储器20。
所述存储器20存储有待识别的图像文件和/或视频文件。其中,所述存储器20中存储的图像文件和/或视频文件为预先存储于存储器20中的,也可为临时从服务器下载或者从其他电子装置100接收的。
在一些实施例中,所述电子装置100还包括通信单元30,所述处理器10可预先通过通信单元30与服务器或其他电子装置建立通信连接,而从服务器或其他电子装置接收待识别的图像文件和/或视频文件,并存储于所述存储器20中。所述通信单元30可为有线或无线通信模块,例如可为有线网络接口单元、WIFI模组、蓝牙模组等。
所述处理器10用于对所述存储器20存储的待识别的图像文件和/或视频文件进行分析,识别是否为360度全景图像和/或360度全景视频。
其中,所述处理器10用于执行如图1及图3-5所示的任一方法来识别待识别的图像文件是否为360度全景图像。
所述处理器10并用于执行如图6-7所示的任一方法步骤及图3-5所示的相关方法步骤来识别待识别的视频文件是否为360度全景视频。
在一些实施例中,所述存储器20中存储有若干程序指令,所述处理器10调用执行所述若干程序指令后,执行如图1及图3-5所示的任一方法来识别待识别的图像文件是否为360度全景图像,和/或执行如图6-7所示的任一方法步骤及图3-5所示的相关方法步骤来识别待识别的视频文件是否为360度全景视频。
例如,所述处理器10执行如下方法识别待识别的图像文件是否为360度 全景图像:获取待识别的图像文件;根据待识别的图像文件确定目标图像帧;判断目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于一第一阈值,以及判断目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值;如果是,确定所述待识别的图像文件为360度全景图像;如果否,确定所述待识别的图像文件不为360度全景图像。
又例如,所述处理器10执行如下方法识别待识别的视频文件是否为360度全景视频:获取待识别的视频文件;从所述待识别的视频文件中提取若干代表图像帧;确定每一代表图像帧是否为360度全景图像;确定为360度全景图像的代表图像帧所占的比例是否大于预设比例;如果是,确定所述视频文件为360度全景视频,如果否,确定所述视频文件不为360度全景视频。
其中,所述处理器40可为微控制器、微处理器、单片机、数字信号处理器等。
所述存储器20可为存储卡、固态存储器、微硬盘、光盘等任意可存储信息的存储设备。
在一些实施例中,本发明还提供一种计算机可读存储介质,所述计算机可读存储介质中存储有若干程序指令,所述若干程序指令供处理器10调用执行后,执行图1、图3-7的任一方法步骤,从而识别图像文件是否为360度全景图像和/或识别视频文件是否为360度全景视频。在一些实施例中,所述计算机存储介质即为所述存储器20,可为存储卡、固态存储器、微硬盘、光盘等任意可存储信息的存储设备。
所述电子装置100可为手机、平板电脑、笔记本电脑、桌面型电脑等,也可为智能头盔,智能眼镜等头戴式设备。
从而,通过本发明的360度全景图像及360度全景视频的识别方法及电子装置,通过确定目标图像帧,然后分析目标图像帧的第一列的像素点与最后一列的像素点之间的灰度差值以及分析目标图像帧的第一行的像素点之间的灰度差值以及最后一行的像素点之间的灰度差值来判断目标图像帧是否为360度全景图像帧,继而判断对应的图像或视频是否为360度全景图像或360度全景视频,可有效提高识别准确度。
以上所述是本发明的优选实施例,应当指出,对于本技术领域的普通技术人员来说,在不脱离本发明原理的前提下,还可以做出若干改进和润饰,这些改进和润饰也视为本发明的保护范围。

Claims (20)

  1. 一种360度全景图像的识别方法,其特征在于,所述识别方法包括:
    获取待识别的图像文件;
    根据待识别的图像文件确定目标图像帧;
    判断所述目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断所述目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值;
    如果是,确定所述待识别的图像文件为360度全景图像,否则,确定所述待识别的图像文件不为360度全景图像。
  2. 如权利要求1所述的识别方法,其特征在于,所述步骤“判断所述目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断所述目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值”包括:
    获取所述目标图像帧的第一列的像素点的灰度值以及最后一列的像素点的灰度值;
    计算第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度;
    判断所述匹配度是否大于或等于第一阈值;
    获取所述目标图像帧的第一行的所有像素点的灰度值以及最后一行的所有像素点的灰度值;
    计算所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差;
    判断所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值。
  3. 如权利要求1所述的识别方法,其特征在于,所述步骤“根据待识别的图像文件确定目标图像帧”包括:
    判断所述待识别的图像为三维图像还是二维图像;
    如果为三维图像,提取所述待识别的图像中的左眼图像或右眼图像,确定所述提取出来的左眼图像或右眼图像为目标图像帧;
    如果为二维图像,确定所述待识别的图像为所述目标图像帧。
  4. 如权利要求2所述的识别方法,其特征在于,所述步骤“计算第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度”包括:
    依次以第一列和/或最后一列的每一个像素点为中心点,获取所述第一列和/或最后一列的当前为中心点的像素点上下相邻的预设个像素点的灰度值以及所述当前为中心点的像素点的像素值;
    计算所述获取到的包括当前为中心点的像素点在内的若干像素点的灰度值的平均值,将所述平均值作为所述当前为中心点的像素点的校正灰度值,以此一一得出所述第一列和最后一列的每一个像素点的校正灰度值;
    分别计算第一列的每一个像素点的校正灰度值与最后一列对应行的像素点的校正灰度值的差值而得出若干个灰度值差值;以及
    计算所述若干个灰度值差值中小于或等于预设差值的灰度值差值所占的比例,将所述比例确定为所述匹配度。
  5. 如权利要求2所述的识别方法,其特征在于,所述步骤“计算所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差”包括:
    计算所述第一行的所有像素点的灰度值的方差得到所有像素点的灰度值之差,以及计算所述最后一行的所有像素点的灰度值的方差得到所述最后一行的所有像素点的灰度值之差。
  6. 一种360度全景视频的识别方法,其特征在于,所述识别方法包括:
    获取待识别的视频文件;
    从所述待识别的视频文件中提取若干代表图像帧;
    确定每一代表图像帧是否为360度全景图像;
    确定为360度全景图像的代表图像帧所占的比例是否大于预设比例;
    如果是,确定所述视频文件为360度全景视频,如果否,确定所述视频文件不为360度全景视频。
  7. 如权利要求6所述的识别方法,其特征在于,所述步骤“确定每一代表图像帧是否为360度全景图像”包括:
    确定每一代表图像帧对应的目标图像帧;
    判断每一目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断每一目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值;
    如果是,确定对应的代表图像帧为360度全景图像,否则,确定对应的代表图像帧不为360度全景图像。
  8. 如权利要求7所述的识别方法,其特征在于,所述步骤“判断每一目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断每一目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值”包括:
    获取所述目标图像帧的第一列的像素点的灰度值以及最后一列的像素点的灰度值;
    计算第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度;
    判断所述匹配度是否大于或等于第一阈值;
    获取所述目标图像帧的第一行的所有像素点的灰度值以及最后一行的所有像素点的灰度值;
    计算所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差;
    判断所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值。
  9. 如权利要求7所述的识别方法,其特征在于,所述步骤“确定每一代表图像帧对应的目标图像帧”包括:
    判断所述代表图像帧为三维图像还是二维图像;
    如果为三维图像,提取所述代表图像帧中的左眼图像或右眼图像,确定所 述提取出来的左眼图像或右眼图像为目标图像帧;
    如果为二维图像,确定所述代表图像帧为所述目标图像帧。
  10. 如权利要求8所述的识别方法,其特征在于,所述步骤“计算第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度”包括:
    依次以第一列和/或最后一列的每一个像素点为中心点,获取所述第一列和/或最后一列的当前为中心点的像素点上下相邻的预设个像素点的灰度值以及所述当前为中心点的像素点的像素值;
    计算所述获取到的包括当前为中心点的像素点在内的若干像素点的灰度值的平均值,将所述平均值作为所述当前为中心点的像素点的校正灰度值,以此一一得出所述第一列和最后一列的每一个像素点的校正灰度值;
    分别计算第一列的每一个像素点的校正灰度值与最后一列对应行的像素点的校正灰度值的差值而得出若干个灰度值差值;以及
    计算所述若干个灰度值差值中小于或等于预设差值的灰度值差值所占的比例,将所述比例确定为所述匹配度。
  11. 如权利要求8所述的识别方法,其特征在于,所述步骤“计算所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差”包括:
    计算所述第一行的所有像素点的灰度值的方差得到所有像素点的灰度值之差,以及计算所述最后一行的所有像素点的灰度值的方差得到所述最后一行的所有像素点的灰度值之差。
  12. 如权利要求6所述的识别方法,其特征在于,所述步骤“从所述待识别的视频文件中提取若干代表图像帧”包括:
    选取视频文件中的关键帧、或选取画面变化丰富的帧图像,或根据多个代表图像帧覆盖视频文件的时间段超过预设值来选取多个代表图像帧。
  13. 一种电子装置,包括处理器以及存储器,所述存储器中存储有待识别的视频文件,其特征在于,所述处理器用于:
    获取待识别的视频文件;
    从所述待识别的视频文件中提取若干代表图像帧;
    确定每一代表图像帧是否为360度全景图像;
    确定为360度全景图像的代表图像帧所占的比例是否大于预设比例;
    如果是,确定所述视频文件为360度全景视频,如果否,确定所述视频文件不为360度全景视频。
  14. 如权利要求13所述的电子装置,其特征在于,所述处理器确定每一代表图像帧是否为360度全景图像,包括:
    确定每一代表图像帧对应的目标图像帧;
    判断每一目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断每一目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值;
    如果是,确定对应的代表图像帧为360度全景图像,否则,确定对应的代表图像帧不为360度全景图像。
  15. 如权利要求14所述的电子装置,其特征在于,所述处理器判断每一目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断每一目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值,包括:
    获取所述目标图像帧的第一列的像素点的灰度值以及最后一列的像素点的灰度值;
    计算第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度;
    判断所述匹配度是否大于或等于第一阈值;
    获取所述目标图像帧的第一行的所有像素点的灰度值以及最后一行的所有像素点的灰度值;
    计算所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差;
    判断所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值。
  16. 如权利要求14所述的电子装置,其特征在于,所述处理器确定每一 代表图像帧对应的目标图像帧,包括:
    判断所述代表图像帧为三维图像还是二维图像;
    如果为三维图像,提取所述代表图像帧中的左眼图像或右眼图像,确定所述提取出来的左眼图像或右眼图像为目标图像帧;
    如果为二维图像,确定所述代表图像帧为所述目标图像帧。
  17. 如权利要求15所述的电子装置,其特征在于,所述处理器计算第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度,包括:
    依次以第一列和/或最后一列的每一个像素点为中心点,获取所述第一列和/或最后一列的当前为中心点的像素点上下相邻的预设个像素点的灰度值以及所述当前为中心点的像素点的像素值;
    计算所述获取到的包括当前为中心点的像素点在内的若干像素点的灰度值的平均值,将所述平均值作为所述当前为中心点的像素点的校正灰度值,以此一一得出所述第一列和最后一列的每一个像素点的校正灰度值;
    分别计算第一列的每一个像素点的校正灰度值与最后一列对应行的像素点的校正灰度值的差值而得出若干个灰度值差值;以及
    计算所述若干个灰度值差值中小于或等于预设差值的灰度值差值所占的比例,将所述比例确定为所述匹配度。
  18. 如权利要求15所述的电子装置,其特征在于,所述处理器计算所述第一行的所有像素点的灰度值之差以及所述最后一行的所有像素点的灰度值之差,包括:
    计算所述第一行的所有像素点的灰度值的方差得到所有像素点的灰度值之差,以及计算所述最后一行的所有像素点的灰度值的方差得到所述最后一行的所有像素点的灰度值之差。
  19. 如权利要求15所述的电子装置,其特征在于,所述处理器通过选取视频文件中的关键帧作为代表图像帧、或选取画面变化丰富的帧图像作为代表图像帧,或选取覆盖视频文件的时间段超过预设范围的多个图像帧作为所述多个代表图像帧。
  20. 一种电子装置,包括处理器以及存储器,所述存储器中存储有待识别的图像文件,其特征在于,所述处理器用于:
    获取待识别的图像文件;
    根据待识别的图像文件确定目标图像帧;
    判断所述目标图像帧的第一列的像素点的灰度值与最后一列的像素点的灰度值之间的匹配度是否大于或等于第一阈值,以及判断所述目标图像帧的第一行的所有像素点的灰度值之差以及最后一行的所有像素点的灰度值之差是否均小于或等于第二阈值;
    如果是,确定所述待识别的图像文件为360度全景图像,否则,确定所述待识别的图像文件不为360度全景图像。
PCT/CN2017/087475 2017-06-07 2017-06-07 360度全景图像及360度全景视频的识别方法及电子装置 Ceased WO2018223318A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2017/087475 WO2018223318A1 (zh) 2017-06-07 2017-06-07 360度全景图像及360度全景视频的识别方法及电子装置
CN201780004657.7A CN108475432B (zh) 2017-06-07 2017-06-07 360度全景图像及360度全景视频的识别方法及电子装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/087475 WO2018223318A1 (zh) 2017-06-07 2017-06-07 360度全景图像及360度全景视频的识别方法及电子装置

Publications (1)

Publication Number Publication Date
WO2018223318A1 true WO2018223318A1 (zh) 2018-12-13

Family

ID=63266498

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/087475 Ceased WO2018223318A1 (zh) 2017-06-07 2017-06-07 360度全景图像及360度全景视频的识别方法及电子装置

Country Status (2)

Country Link
CN (1) CN108475432B (zh)
WO (1) WO2018223318A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103581644A (zh) * 2012-07-31 2014-02-12 索尼公司 图像处理装置、图像处理方法及程序
CN103950410A (zh) * 2014-04-29 2014-07-30 深圳前向启创数码技术有限公司 全景式辅助驾驶方法及系统
CN106162146A (zh) * 2016-07-29 2016-11-23 暴风集团股份有限公司 自动识别并播放全景视频的方法及系统
CN106327531A (zh) * 2016-08-18 2017-01-11 成都虚拟世界科技有限公司 全景视频的识别方法及设备、播放视频方法及设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106331848B (zh) * 2016-08-18 2019-04-19 成都虚拟世界科技有限公司 全景视频的识别方法及设备、播放视频方法及设备
CN106534974B (zh) * 2016-12-16 2019-04-26 暴风集团股份有限公司 自动识别立方体全景视频的方法及系统
CN106803994B (zh) * 2017-03-29 2019-07-26 暴风集团股份有限公司 识别四棱锥全景视频的方法及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103581644A (zh) * 2012-07-31 2014-02-12 索尼公司 图像处理装置、图像处理方法及程序
CN103950410A (zh) * 2014-04-29 2014-07-30 深圳前向启创数码技术有限公司 全景式辅助驾驶方法及系统
CN106162146A (zh) * 2016-07-29 2016-11-23 暴风集团股份有限公司 自动识别并播放全景视频的方法及系统
CN106327531A (zh) * 2016-08-18 2017-01-11 成都虚拟世界科技有限公司 全景视频的识别方法及设备、播放视频方法及设备

Also Published As

Publication number Publication date
CN108475432A (zh) 2018-08-31
CN108475432B (zh) 2021-12-28

Similar Documents

Publication Publication Date Title
CN108024107B (zh) 图像处理方法、装置、电子设备及计算机可读存储介质
JP4505362B2 (ja) 赤目検出装置および方法並びにプログラム
US10027878B2 (en) Detection of object in digital image
KR102277048B1 (ko) 미리보기 사진 블러링 방법 및 장치 및 저장 매체
US8599270B2 (en) Computing device, storage medium and method for identifying differences between two images
EP4013033A1 (en) Method and apparatus for focusing on subject, and electronic device, and storage medium
CN107622497B (zh) 图像裁剪方法、装置、计算机可读存储介质和计算机设备
CN109510946B (zh) Hdr场景检测方法及系统
WO2017215527A1 (zh) Hdr场景侦测方法、装置和计算机存储介质
CN107172354A (zh) 视频处理方法、装置、电子设备及存储介质
WO2020093653A1 (zh) 色彩调整方法、色彩调整装置、电子设备及计算机可读存储介质
WO2023151210A1 (zh) 图像处理方法、电子设备及计算机可读存储介质
US8582813B2 (en) Object detection device which detects object based on similarities in different frame images, and object detection method and computer-readable medium recording program
US11315360B2 (en) Live facial recognition system and method
CN107995476B (zh) 一种图像处理方法及装置
US9832395B2 (en) Information processing method applied to an electronic device and electronic device having at least two image capturing units that have the same image capturing direction
JP6374849B2 (ja) ユーザ端末、色彩補正システム及び色彩補正方法
CN108737797B (zh) 白平衡处理方法、装置和电子设备
US12035033B2 (en) DNN assisted object detection and image optimization
US8824734B2 (en) Device, method and recording to determine depressed portions of image regions using shortcut line analysis
JP6217225B2 (ja) 画像照合装置、画像照合方法及びプログラム
CN108805883B (zh) 一种图像分割方法、图像分割装置及电子设备
CN108475432B (zh) 360度全景图像及360度全景视频的识别方法及电子装置
US8538142B2 (en) Face-detection processing methods, image processing devices, and articles of manufacture
JP7709053B2 (ja) 情報処理装置、情報処理システム、制御方法、プログラム

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: 17912697

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: 17912697

Country of ref document: EP

Kind code of ref document: A1