WO2017101124A1 - 一种监控系统中背景识别的方法及系统 - Google Patents

一种监控系统中背景识别的方法及系统 Download PDF

Info

Publication number
WO2017101124A1
WO2017101124A1 PCT/CN2015/097975 CN2015097975W WO2017101124A1 WO 2017101124 A1 WO2017101124 A1 WO 2017101124A1 CN 2015097975 W CN2015097975 W CN 2015097975W WO 2017101124 A1 WO2017101124 A1 WO 2017101124A1
Authority
WO
WIPO (PCT)
Prior art keywords
picture
pictures
background
frame
area
Prior art date
Application number
PCT/CN2015/097975
Other languages
English (en)
French (fr)
Inventor
张北江
Original Assignee
张北江
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 张北江 filed Critical 张北江
Priority to PCT/CN2015/097975 priority Critical patent/WO2017101124A1/zh
Priority to CN201580001326.9A priority patent/CN105830437B/zh
Publication of WO2017101124A1 publication Critical patent/WO2017101124A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/183Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/40Scenes; Scene-specific elements in video content
    • G06V20/41Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items

Definitions

  • the present invention relates to the field of image and monitoring, and in particular, to a method and system for background recognition in a monitoring system.
  • Video surveillance is an important part of the security system, English Cameras and Surveillance.
  • Traditional surveillance systems include front-end cameras, transmission cables, and video surveillance platforms.
  • the camera can be divided into network digital cameras and analog cameras, which can be used as the front-end video image signal acquisition. It is a comprehensive system with strong defense capabilities.
  • Video surveillance is widely used in many applications because of its intuitiveness, accuracy, and ambiguity.
  • the latest surveillance system can use the smartphone to automatically identify, store and automatically alert images.
  • the video data is transmitted back to the control panel through 3G/4G/WIFI (also can be used as a smart phone), and the host can perform operations such as viewing, recording, playback, recall and storage of images. Thereby achieving video surveillance of mobile internet.
  • 3G/4G/WIFI also can be used as a smart phone
  • the present application provides a method and system for background recognition in a monitoring system. It can improve the reliability of background recognition.
  • a method of monitoring a background in a system comprising the steps of: [0006] acquiring a multi-frame picture in sequence, comparing the multi-frame picture with the pre-stored original background picture, and removing the original background picture to obtain an initial processed picture of the multi-frame;
  • the acquiring the multiple frames of the picture in sequence includes:
  • sampling all the pictures comparing the sampled pictures to determine the similarity of the sampled pictures, and determining the pictures of all the frames between the sampled pictures whose similarities are within the set interval range as the acquired multiple frames. image.
  • the first processed image of the multi-frame is compared to obtain the image data of the same area; [0012] the area in the picture whose color difference is within the set range is used as the picture data of the same area.
  • a system for monitoring background recognition in a system comprising:
  • an obtaining unit configured to sequentially acquire a multi-frame picture
  • a pre-processing unit configured to compare the multi-frame picture with the pre-stored original background picture, and remove the original background picture to obtain an initial processed picture of the multi-frame
  • a calculating unit configured to compare the initial processed pictures of the multi-frame to obtain the picture data of the same area, and superimpose the picture data of the same area of the multiple frames, and calculate the superposed area;
  • a background determining unit configured to calculate a ratio of a superimposed area to a minimum area in a picture of the same area in the multi-frame initial processing picture, and if the ratio is greater than a set threshold, determine to be a non-background, if the ratio is less than Set the threshold, then determine the background, delete.
  • the acquiring unit is specifically configured to:
  • sampling all the pictures comparing the sampled pictures to determine the similarity of the sampled pictures, and determining the pictures of all the frames between the sampled pictures whose similarities are within the set interval range as the acquired multi-frames. image.
  • the calculating unit is specifically configured to use, as the picture data of the same area, an area in which a color difference in the picture is within a set range.
  • the technical solution provided by the present invention identifies a dynamic background after removing the static background.
  • the first preferred embodiment of the present invention can effectively reduce the range of the acquired multi-frame picture by sampling. Calculate the amount and improve the accuracy of background recognition.
  • FIG. 1 is a flow chart of a method for background recognition in a monitoring system according to a first preferred embodiment of the present invention
  • FIG. 2 is a structural diagram of a system for background recognition in a monitoring system according to a second preferred embodiment of the present invention.
  • FIG. 1 is a schematic diagram of a background recognition method in a monitoring system according to an embodiment of the present invention.
  • the method is implemented by a monitoring system, including but not limited to, a building monitoring system, a computer, and an intelligent terminal.
  • the device, the method shown in Figure 1 includes the following steps:
  • Step S101 Obtain a multi-frame picture in sequence, compare the multi-frame picture with the pre-stored original background picture, and remove the original background picture to obtain an initial processed picture of the multi-frame; [0028] In the above step S101, it is necessary to obtain a multi-frame picture in order, because for the picture data, if the picture is acquired at a fixed inter-turn interval, the picture may be sampled. And then comparing the sampled pictures to determine the similarity of the sampled pictures, and determining, for the pictures of all the frames between the sampled pictures whose similarities are within the set interval, as the acquired multi-frame pictures, the method may first be reduced.
  • the above method of aligning and removing the original background picture may employ a prior art method.
  • Step S102 Perform image data of the same area by comparing the initial processed pictures of the multiple frames, and superimpose the picture data of the same area of the multiple frames, and calculate the superposed area;
  • the method for obtaining the picture data of the same area in the foregoing comparison may be various.
  • the color difference of the picture data of the same area may be compared.
  • the way to obtain the picture data of the same area in a practical example, for the same area, since the inter-turn interval of the multi-frame picture is very short, assuming that there is a person passing through, then the color difference of the human skin is not different under the same conditions. Large, so that the same area can be marked by the identification of the color difference.
  • multiple regions of the multi-frame picture can be obtained by comparing with the original background picture. Then, the areas of the same area of the plurality of frames are compared, and the areas having substantially the same area and the same relative positional relationship in the picture are set as the same area.
  • Step S103 calculating a ratio of the superimposed area to a minimum area in a picture of the same area in the multi-frame initial processing picture, and if the ratio is greater than a set threshold, determining to be a non-background, if the ratio is less than a set threshold, Then determine as background, delete.
  • the method for superimposing pictures in the above step S103 may adopt a method of the prior art, and the first preferred embodiment of the present invention does not limit the specific implementation manner of the above superposition.
  • a second preferred embodiment of the present invention provides a system for monitoring a background in a system, the system comprising:
  • the obtaining unit 201 is configured to acquire multiple frames of pictures in sequence
  • the acquisition unit needs to have a method for acquiring multiple frames in order, because for the image data, if the image is acquired at a fixed inter-turn interval, the image may be less accurate. Then, the sampled pictures are compared to determine the similarity of the sampled pictures, and the pictures of all the frames between the sampled pictures whose similarities are within the set interval range are determined as the acquired multi-frame pictures, which can first reduce the calculation. Quantity, because it is obtained by sampling the picture, because only the sampled picture is processed to determine the range of acquiring the multi-frame picture, and in addition, since it is sampling, it is not easy to miss the corresponding picture, so it has the identification High accuracy and small calculation.
  • the pre-processing unit 202 is configured to compare the multi-frame picture with the pre-stored original background picture, and remove the original background picture to obtain an initial processed picture of the multi-frame;
  • the calculating unit 203 is configured to compare the initial processed pictures of the multiple frames to obtain the picture data of the same area, and superimpose the picture data of the same area of the multiple frames, and calculate the superposed area;
  • the method for obtaining the picture data of the same area in the foregoing comparison may be various.
  • the color difference of the picture data of the same area may be compared.
  • the way to obtain the picture data of the same area in a practical example, for the same area, since the inter-turn interval of the multi-frame picture is very short, assuming that there is a person passing through, then the color difference of the human skin is not different under the same conditions. Large, so that the same area can be marked by the identification of the color difference.
  • multiple regions of the multi-frame picture can be obtained by comparing with the original background picture. Then, the areas of the same area of the plurality of frames are compared, and the areas having substantially the same area and the same relative positional relationship in the picture are set as the same area.
  • the background determining unit 204 is configured to calculate a ratio of the area after the superimposition to the minimum area of the same area in the multi-frame initial processing picture, and if the ratio is greater than the set threshold, determine that the background is non-background, if the ratio is smaller than Set the threshold, then determine the background, delete.
  • the method for superimposing pictures in the background determining unit 204 may adopt a method of the prior art, and the first preferred embodiment of the present invention does not limit the specific implementation of the superimposing.
  • the acquiring unit is specifically configured to:
  • sampling all the pictures comparing the sampled pictures to determine the similarity of the sampled pictures, and determining the pictures of all the frames between the sampled pictures whose similarities are within the set interval range as the acquired multi-frames. image.
  • the calculating unit is specifically configured to use, as the picture data of the same area, an area in the picture whose color difference is within a set range.
  • a program instructing related hardware can be stored in a computer readable storage medium, the storage medium. It can include: flash drive, read-only memory (English: Read-Only Memory, ROM for short), random access memory (English: Random Access Memory, RAM for short), disk or CD.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

本发明公开了一种监控系统中背景识别的方法及系统,所述方法包括如下步骤:按顺序获取多帧图片,将多帧图片与预先存储的原始背景图片比对去掉原始背景图片得到多帧的初始处理图片;将多帧的初始处理图片进行比对获取相同区域的图片数据,将多帧的相同区域的图片数据叠加后,计算叠加后的面积;计算叠加后的面积与多帧初始处理图片中相同区域的图片中最小面积的比值,如该比值大于设定阈值,则确定为非背景,如该比值小于设定阈值,则确定为背景,删除。本发明提供的技术方案具有背景识别精度高的优点。

Description

一种监控系统中背景识别的方法及系统 技术领域
[0001] 本发明涉及图像及监控领域, 尤其涉及一种监控系统中背景识别的方法及系统 背景技术
[0002] 视频监控是安全防范系统的重要组成部分, 英文 Cameras and Surveillance。 传 统的监控系统包括前端摄像机、 传输线缆、 视频监控平台。 摄像机可分为网络 数字摄像机和模拟摄像机, 可作为前端视频图像信号的采集。 它是一种防范能 力较强的综合系统。 视频监控以其直观、 准确、 及吋和信息内容丰富而广泛应 用于许多场合。 近年来, 随着计算机、 网络以及图像处理、 传输技术的飞速发 展, 视频监控技术也有了长足的发展。 最新的监控系统可以使用智能手机担当 , 同吋对图像进行自动识别、 存储和自动报警。 视频数据通过 3G/4G/WIFI传回 控制主机 (也可以是智能手机担当) , 主机可对图像进行实吋观看、 录入、 回 放、 调出及储存等操作。 从而实现移动互联的视频监控。
[0003] 现有的视频监控均带有自动报警功能, 现有的自动报警功能多数是对活动物体 进行识别, 然后对相应的情况进行报警, 活动物体的识别是基于背景识别技术 的, 现有的背景识别技术通常是通过多帧图片比对的方式来确定静态物体, 从 而对背景进行识别和剔除, 但是在实际应用中发现, 现有技术的技术方案因为 仅仅是静态物体的识别, 无法完全进行背景识别消除, 所以其具有背景识别不 准确的缺点。
技术问题
[0004] 本申请提供一种监控系统中背景识别的方法及系统。 可以提高背景识别的可靠 性。
问题的解决方案
技术解决方案
[0005] 一方面, 提供监控系统中背景识别的方法, 所述方法包括如下步骤: [0006] 按顺序获取多帧图片, 将多帧图片与预先存储的原始背景图片比对去掉原始背 景图片得到多帧的初始处理图片;
[0007] 将多帧的初始处理图片进行比对获取相同区域的图片数据, 将多帧的相同区域 的图片数据叠加后, 计算叠加后的面积;
[0008] 计算叠加后的面积与多帧初始处理图片中相同区域的图片中最小面积的比值, 如该比值大于设定阈值, 则确定为非背景, 如该比值小于设定阈值, 则确定为 背景, 刪除。
[0009] 可选的, 所述按顺序获取多帧图片具体, 包括:
[0010] 对所有图片进行采样, 对采样后的图片进行比对确定采样图片的相似度, 对相 似度在设定区间范围内的采样图片之间的所有的帧的图片确定为获取的多帧图 片。
[0011] 可选的, 所述将多帧的初始处理图片进行比对获取相同区域的图片数据; [0012] 将图片中色差在设定范围内的区域作为相同区域的图片数据。
[0013] 另一方面, 提供一种监控系统中背景识别的系统, 所述系统包括:
[0014] 获取单元, 用于按顺序获取多帧图片;
[0015] 预处理单元, 用于将多帧图片与预先存储的原始背景图片比对去掉原始背景图 片得到多帧的初始处理图片;
[0016] 计算单元, 用于将多帧的初始处理图片进行比对获取相同区域的图片数据, 将 多帧的相同区域的图片数据叠加后, 计算叠加后的面积;
[0017] 背景确定单元, 用于计算叠加后的面积与多帧初始处理图片中相同区域的图片 中最小面积的比值, 如该比值大于设定阈值, 则确定为非背景, 如该比值小于 设定阈值, 则确定为背景, 刪除。
[0018] 可选的, 所述获取单元, 具体用于
[0019] 对所有图片进行采样, 对采样后的图片进行比对确定采样图片的相似度, 对相 似度在设定区间范围内的采样图片之间的所有的帧的图片确定为获取的多帧图 片。
[0020] 可选的, 所述计算单元, 具体用于将图片中色差在设定范围内的区域作为相同 区域的图片数据。 发明的有益效果
有益效果
[0021] [0005]本发明提供的技术方案对静态背景去除以后, 还对动态的背景进行识别
, 并且通过多图片叠加的方式对动态背景进行去除, 所以其具有识别效果好的 优点, 另外, 本发明第一较佳实施方式通过采样的方式来确定获取的多帧图片 的范围能够有效的降低计算量, 提高背景识别的精度。
对附图的简要说明
附图说明
[0022] 图 1为本发明第一较佳实施方式提供的一种监控系统中背景识别的方法的流程 图;
[0023] 图 2为本发明第二较佳实施方式提供的一种监控系统中背景识别的系统的结构 图。 本发明的实施方式
[0024] [0008]为了更清楚地说明本发明实施例的技术方案, 下面将对实施例描述中所 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。
[0025] 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作出创造性劳动前 提下所获得的所有其他实施例, 都属于本发明保护的范围。
[0026] 请参考图 1, 图 1是本发明实施例提出的一种监控系统中背景识别的方法, 该方 法由监控系统执行, 该监控系统包括但不限于, 楼宇监控系统、 计算机、 智能 终端等设备, 该方法如图 1所示, 包括如下步骤:
[0027] 步骤 S101、 按顺序获取多帧图片, 将多帧图片与预先存储的原始背景图片比对 去掉原始背景图片得到多帧的初始处理图片; [0028] 上述步骤 S101中按顺序获取多帧图片需要有一个获取的方式, 因为对于图片数 据来说, 如按固定的吋间间隔获取图片可能会不太准确, 这里通过对图片的进 行采样处理, 然后对采样后的图片进行比对确定采样图片的相似度, 对相似度 在设定区间范围内的采样图片之间的所有的帧的图片确定为获取的多帧图片, 此方式首先可以降低计算量, 因为其是通过采样图片来获取的, 因为仅仅对采 样的图片进行处理来确定获取多帧图片的范围, 另外, 由于是采样, 这样也不 太容易漏掉相应的图片, 所以其具有识别准确度高, 计算量小的优点。
[0029] 上述比对和去掉原始背景图片的方式可以采用现有技术的方法。
[0030] 步骤 S102、 将多帧的初始处理图片进行比对获取相同区域的图片数据, 将多帧 的相同区域的图片数据叠加后, 计算叠加后的面积;
[0031] 上述比对获取相同区域的图片数据的方法可以有多种, 例如, 在本发明第一较 佳实施方式的一个实施例中, 可以采用对相同区域的图片数据的色差进行比对 的方式来获取相同区域的图片数据, 以一个实际例子来说, 对于相同区域, 由 于多帧图片的吋间间隔非常短, 假设此吋有个人经过, 那么人类的皮肤的色差 在相同条件下差别不大, 这样就可以通过这样色差的识别即可以对相同区域进 行标注, 当然在实际应用中还可以采用其他的方式, 例如通过与原始背景图片 比对即可以获取上述多帧图片的多个区域, 然后将多个帧的相同区域的面积进 行比对, 基本相同的面积且在图片中相对位置关系也相同的区域设定为相同区 域。
[0032] 步骤 S103、 计算叠加后的面积与多帧初始处理图片中相同区域的图片中最小面 积的比值, 如该比值大于设定阈值, 则确定为非背景, 如该比值小于设定阈值 , 则确定为背景, 刪除。
[0033] 上述步骤 S103中图片叠加的方法可以采用现有技术的方法, 本发明第一较佳实 施方式并不限定上述叠加的具体实现方式。
[0034] 本发明第一较佳实施方式提供的技术方案对静态背景去除以后, 还对动态的背 景进行识别, 并且通过多图片叠加的方式对动态背景进行去除, 所以其具有识 别效果好的优点, 另外, 本发明第一较佳实施方式通过采样的方式来确定获取 的多帧图片的范围能够有效的降低计算量, 提高背景识别的精度。 [0035] 另一方面, 本发明第二较佳实施方式提供一种监控系统中背景识别的系统, 所 述系统包括:
[0036] 获取单元 201, 用于按顺序获取多帧图片;
[0037] 获取单元中按顺序获取多帧图片需要有一个获取的方式, 因为对于图片数据来 说, 如按固定的吋间间隔获取图片可能会不太准确, 这里通过对图片的进行采 样处理, 然后对采样后的图片进行比对确定采样图片的相似度, 对相似度在设 定区间范围内的采样图片之间的所有的帧的图片确定为获取的多帧图片, 此方 式首先可以降低计算量, 因为其是通过采样图片来获取的, 因为仅仅对采样的 图片进行处理来确定获取多帧图片的范围, 另外, 由于是采样, 这样也不太容 易漏掉相应的图片, 所以其具有识别准确度高, 计算量小的优点。
[0038] 预处理单元 202, 用于将多帧图片与预先存储的原始背景图片比对去掉原始背 景图片得到多帧的初始处理图片;
[0039] 计算单元 203, 用于将多帧的初始处理图片进行比对获取相同区域的图片数据 , 将多帧的相同区域的图片数据叠加后, 计算叠加后的面积;
[0040] 上述比对获取相同区域的图片数据的方法可以有多种, 例如, 在本发明第一较 佳实施方式的一个实施例中, 可以采用对相同区域的图片数据的色差进行比对 的方式来获取相同区域的图片数据, 以一个实际例子来说, 对于相同区域, 由 于多帧图片的吋间间隔非常短, 假设此吋有个人经过, 那么人类的皮肤的色差 在相同条件下差别不大, 这样就可以通过这样色差的识别即可以对相同区域进 行标注, 当然在实际应用中还可以采用其他的方式, 例如通过与原始背景图片 比对即可以获取上述多帧图片的多个区域, 然后将多个帧的相同区域的面积进 行比对, 基本相同的面积且在图片中相对位置关系也相同的区域设定为相同区 域。
[0041] 背景确定单元 204, 用于计算叠加后的面积与多帧初始处理图片中相同区域的 图片中最小面积的比值, 如该比值大于设定阈值, 则确定为非背景, 如该比值 小于设定阈值, 则确定为背景, 刪除。
[0042] 上述背景确定单元 204中图片叠加的方法可以采用现有技术的方法, 本发明第 一较佳实施方式并不限定上述叠加的具体实现方式。 [0043] 可选的, 所述获取单元, 具体用于
[0044] 对所有图片进行采样, 对采样后的图片进行比对确定采样图片的相似度, 对相 似度在设定区间范围内的采样图片之间的所有的帧的图片确定为获取的多帧图 片。
[0045] 可选的, 所述计算单元, 具体用于将图片中色差在设定范围内的区域作为相同 区域的图片数据。
[0046] 需要说明的是, 对于前述的各个方法实施例, 为了简单描述, 故将其都表述为 一系列的动作组合, 但是本领域技术人员应该知悉, 本发明并不受所描述的动 作顺序的限制, 因为依据本发明, 某一些步骤可以采用其他顺序或者同吋进行 。 其次, 本领域技术人员也应该知悉, 说明书中所描述的实施例均属于优选实 施例, 所涉及的动作和模块并不一定是本发明所必须的。
[0047] 在上述实施例中, 对各个实施例的描述都各有侧重, 某个实施例中没有详细描 述的部分, 可以参见其他实施例的相关描述。
[0048] 本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可 以通过程序来指令相关的硬件来完成, 该程序可以存储于一计算机可读存储介 质中, 存储介质可以包括: 闪存盘、 只读存储器 (英文: Read-Only Memory, 简称: ROM) 、 随机存取器 (英文: Random Access Memory , 简称: RAM) 、 磁盘或光盘等。
[0049] 以上对本发明实施例所提供的内容下载方法及相关设备、 系统进行了详细介绍 , 本文中应用了具体个例对本发明的原理及实施方式进行了阐述, 以上实施例 的说明只是用于帮助理解本发明的方法及其核心思想; 同吋, 对于本领域的一 般技术人员, 依据本发明的思想, 在具体实施方式及应用范围上均会有改变之 处, 综上所述, 本说明书内容不应理解为对本发明的限制。

Claims

权利要求书
[权利要求 1] 一种监控系统中背景识别的方法, 其特征在于, 所述方法包括如下步 骤:
按顺序获取多帧图片, 将多帧图片与预先存储的原始背景图片比对去 掉原始背景图片得到多帧的初始处理图片;
将多帧的初始处理图片进行比对获取相同区域的图片数据, 将多帧的 相同区域的图片数据叠加后, 计算叠加后的面积; 计算叠加后的面积与多帧初始处理图片中相同区域的图片中最小面积 的比值, 如该比值大于设定阈值, 则确定为非背景, 如该比值小于设 定阈值, 则确定为背景, 刪除。
[权利要求 2] 根据权利要求 1所述的方法, 其特征在于, 所述按顺序获取多帧图片 具体, 包括:
对所有图片进行采样, 对采样后的图片进行比对确定采样图片的相似 度, 对相似度在设定区间范围内的采样图片之间的所有的帧的图片确 定为获取的多帧图片。
[权利要求 3] 根据权利要求 1所述的方法, 其特征在于, 所述将多帧的初始处理图 片进行比对获取相同区域的图片数据;
将图片中色差在设定范围内的区域作为相同区域的图片数据。
[权利要求 4] 一种监控系统中背景识别的系统, 其特征在于, 所述系统包括: 获取单元, 用于按顺序获取多帧图片;
预处理单元, 用于将多帧图片与预先存储的原始背景图片比对去掉原 始背景图片得到多帧的初始处理图片;
计算单元, 用于将多帧的初始处理图片进行比对获取相同区域的图片 数据, 将多帧的相同区域的图片数据叠加后, 计算叠加后的面积; 背景确定单元, 用于计算叠加后的面积与多帧初始处理图片中相同区 域的图片中最小面积的比值, 如该比值大于设定阈值, 则确定为非背 景, 如该比值小于设定阈值, 则确定为背景, 刪除。
[权利要求 5] 根据权利要求 4所述的系统, 其特征在于, 所述获取单元, 具体用于 对所有图片进行采样, 对采样后的图片进行比对确定采样图片的相似 度, 对相似度在设定区间范围内的采样图片之间的所有的帧的图片确 定为获取的多帧图片。
[权利要求 6] 根据权利要求 4所述的系统, 其特征在于, 所述计算单元, 具体用于 将图片中色差在设定范围内的区域作为相同区域的图片数据。
PCT/CN2015/097975 2015-12-19 2015-12-19 一种监控系统中背景识别的方法及系统 WO2017101124A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2015/097975 WO2017101124A1 (zh) 2015-12-19 2015-12-19 一种监控系统中背景识别的方法及系统
CN201580001326.9A CN105830437B (zh) 2015-12-19 2015-12-19 一种监控系统中背景识别的方法及系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/097975 WO2017101124A1 (zh) 2015-12-19 2015-12-19 一种监控系统中背景识别的方法及系统

Publications (1)

Publication Number Publication Date
WO2017101124A1 true WO2017101124A1 (zh) 2017-06-22

Family

ID=56988059

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/097975 WO2017101124A1 (zh) 2015-12-19 2015-12-19 一种监控系统中背景识别的方法及系统

Country Status (2)

Country Link
CN (1) CN105830437B (zh)
WO (1) WO2017101124A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112184722A (zh) * 2020-09-15 2021-01-05 上海传英信息技术有限公司 图像处理方法、终端及计算机存储介质
CN114785988A (zh) * 2022-04-11 2022-07-22 广东思域信息科技有限公司 一种基于云计算服务的高清视频监控系统及监控方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114022849A (zh) * 2022-01-05 2022-02-08 鑫安利中(北京)科技有限公司 一种基于大数据的企业安全管理方法及系统

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103618846A (zh) * 2013-11-22 2014-03-05 上海安奎拉信息技术有限公司 一种视频分析中抑制光线突然变化影响的背景去除方法
WO2015129501A1 (ja) * 2014-02-25 2015-09-03 株式会社メガチップス 動体検出装置、動体検出方法及び制御プログラム

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103618846A (zh) * 2013-11-22 2014-03-05 上海安奎拉信息技术有限公司 一种视频分析中抑制光线突然变化影响的背景去除方法
WO2015129501A1 (ja) * 2014-02-25 2015-09-03 株式会社メガチップス 動体検出装置、動体検出方法及び制御プログラム

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
HUANG, LEI ET AL.: "Moving Target Detection Research Based on Background Subtraction", SOFTWARE GUIDE, vol. 8, no. 6, 30 June 2009 (2009-06-30), pages 187 - 188, ISSN: 1672-7800 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112184722A (zh) * 2020-09-15 2021-01-05 上海传英信息技术有限公司 图像处理方法、终端及计算机存储介质
CN112184722B (zh) * 2020-09-15 2024-05-03 上海传英信息技术有限公司 图像处理方法、终端及计算机存储介质
CN114785988A (zh) * 2022-04-11 2022-07-22 广东思域信息科技有限公司 一种基于云计算服务的高清视频监控系统及监控方法

Also Published As

Publication number Publication date
CN105830437A (zh) 2016-08-03
CN105830437B (zh) 2019-08-09

Similar Documents

Publication Publication Date Title
US10674066B2 (en) Method for processing image and electronic apparatus therefor
US11024052B2 (en) Stereo camera and height acquisition method thereof and height acquisition system
WO2020094091A1 (zh) 一种图像抓拍方法、监控相机及监控系统
CN109145771B (zh) 一种人脸抓拍方法及装置
CN109086734B (zh) 一种对人眼图像中瞳孔图像进行定位的方法及装置
WO2017054442A1 (zh) 一种图像信息识别处理方法及装置、计算机存储介质
US20180077359A1 (en) Image processing method and apparatus and terminal device
WO2017101126A1 (zh) 一种监控系统的数据存储方法及系统
CN110991287A (zh) 一种实时视频流人脸检测跟踪方法及检测跟踪系统
CN109495689A (zh) 一种拍摄方法、装置、电子设备及存储介质
WO2016197657A1 (zh) 一种拍照处理方法和装置
US10540546B2 (en) Image processing apparatus, control method, and storage medium
CN105430247A (zh) 一种利用摄像设备拍摄照片的方法与设备
WO2017101124A1 (zh) 一种监控系统中背景识别的方法及系统
CN109447022B (zh) 一种镜头类型识别方法及装置
CN109413760B (zh) 一种无线连接方法及设备
US20200401811A1 (en) Systems and methods for target identification in video
CN113379999A (zh) 火情检测方法、装置、电子设备及存储介质
WO2015003446A1 (en) Methods and systems for image recognition
CN113158773B (zh) 一种活体检测模型的训练方法及训练装置
WO2019129041A1 (zh) 一种亮度调节方法、装置、终端及计算机可读存储介质
CN110223320B (zh) 物体检测跟踪方法和检测跟踪装置
CN105467741B (zh) 一种全景拍照方法及终端
WO2017101125A1 (zh) 一种监控系统中的图像管理方法及系统
WO2017049474A1 (zh) 一种拍摄方法以及智能手环

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: 15910593

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 1205 DATED 30/10/2018)

122 Ep: pct application non-entry in european phase

Ref document number: 15910593

Country of ref document: EP

Kind code of ref document: A1