WO2018059125A1 - 一种基于毫米波图像的人体异物检测方法及系统 - Google Patents

一种基于毫米波图像的人体异物检测方法及系统 Download PDF

Info

Publication number
WO2018059125A1
WO2018059125A1 PCT/CN2017/096101 CN2017096101W WO2018059125A1 WO 2018059125 A1 WO2018059125 A1 WO 2018059125A1 CN 2017096101 W CN2017096101 W CN 2017096101W WO 2018059125 A1 WO2018059125 A1 WO 2018059125A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
foreign object
millimeter wave
preset
foreign
Prior art date
Application number
PCT/CN2017/096101
Other languages
English (en)
French (fr)
Inventor
陈寒江
李志权
祁春超
赵术开
Original Assignee
华讯方舟科技有限公司
深圳市太赫兹科技创新研究院
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 华讯方舟科技有限公司, 深圳市太赫兹科技创新研究院 filed Critical 华讯方舟科技有限公司
Priority to EP17854578.6A priority Critical patent/EP3521864A4/en
Priority to US16/336,772 priority patent/US11194073B2/en
Publication of WO2018059125A1 publication Critical patent/WO2018059125A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
    • G01S13/9021—SAR image post-processing techniques
    • G01S13/9029—SAR image post-processing techniques specially adapted for moving target detection within a single SAR image or within multiple SAR images taken at the same time
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V8/00—Prospecting or detecting by optical means
    • G01V8/005—Prospecting or detecting by optical means operating with millimetre waves, e.g. measuring the black losey radiation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/887—Radar or analogous systems specially adapted for specific applications for detection of concealed objects, e.g. contraband or weapons
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
    • G01S13/9021—SAR image post-processing techniques
    • G01S13/9027—Pattern recognition for feature extraction

Definitions

  • the embodiments of the present invention belong to the field of millimeter wave imaging technology, and in particular, to a human body foreign body detection method and system based on millimeter wave images.
  • the millimeter wave has a certain penetration during the propagation process.
  • the millimeter wave imaging technology can safely and conveniently obtain images of metal or non-metal objects under the cover of the clothing, and does not cause ionization damage to the human body, and can replace the traditional X.
  • Radiation security equipment and metal detectors are used in security inspection equipment in public places such as subways, railway stations, and airports.
  • the existing security inspection equipment based on the millimeter wave imaging technology usually directly detects the foreign matter in the human body according to the gradation characteristics of the obtained millimeter wave image of the human body, because the millimeter wave image has a low gray level and low definition. And it has obvious coherent plaque noise. Therefore, it is not ideal to directly detect the foreign matter by using the gradation characteristics of the millimeter wave grayscale image of the human body.
  • Embodiments of the present invention provide a human body foreign matter detecting method and system based on a millimeter wave image, aiming at solving the existing safety inspection device based on the millimeter wave imaging technology, and directly detecting the gray characteristic of the obtained millimeter wave image of the human body. Foreign body in the human body, the effect is not ideal.
  • An embodiment of the present invention provides a method for detecting a foreign body based on a millimeter wave image, the method comprising:
  • the foreign matter image is displayed as a foreign matter detection result.
  • Another aspect of the invention is also a human body foreign body detecting system based on a millimeter wave image, the system comprising:
  • a human grayscale image acquisition module for acquiring a millimeter wave grayscale image of a human body
  • a foreign object region image extracting module configured to extract a foreign object region image in the millimeter wave grayscale image according to a preset foreign matter imaging characteristic
  • a foreign matter image calculation module configured to calculate the foreign object region image according to a preset foreign object image recognition algorithm, and acquire a foreign matter image in the foreign object region image;
  • a display module configured to display the foreign object image as a foreign matter detection result.
  • the millimeter wave gray image of the human body is acquired, and the foreign object region image in the millimeter wave gray image is extracted according to the preset foreign object imaging characteristic, and the foreign object region image is calculated according to the preset foreign object image recognition algorithm.
  • the foreign matter image in the foreign object region image can greatly improve the accuracy of foreign matter detection.
  • FIG. 1 is a human body foreign object detection based on millimeter wave image according to a first embodiment of the present invention.
  • FIG. 2 is a schematic diagram of a millimeter wave grayscale image according to Embodiment 1 of the present invention.
  • FIG. 3 is a schematic diagram of an image of a foreign object region according to Embodiment 1 of the present invention.
  • FIG. 4 is a schematic diagram of a foreign object image according to Embodiment 1 of the present invention.
  • FIG. 5 is a block diagram showing the flow of step S130 according to the second embodiment of the present invention.
  • FIG. 6 is a block diagram showing the basic structure of a human body foreign body detecting system based on millimeter wave images according to Embodiment 3 of the present invention.
  • FIG. 7 is a structural block diagram of a foreign object image calculation module according to Embodiment 4 of the present invention.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • the method for detecting a foreign body based on a millimeter wave image includes:
  • Step S110 Acquire a millimeter wave grayscale image of the human body.
  • the human body can be lifted up by the top of the head or the hands are raised to the position of the same height as the shoulder, or the human hands can be naturally sagged, or other standing postures conforming to the security standards can be used.
  • the embodiment does not specifically limit the standing posture of the human body, and then uses millimeter wave data acquisition devices (for example, millimeter wave transceivers) to acquire millimeter wave data of the front or back of the human body, and utilizes a millimeter wave imaging system (for example, a millimeter wave imager).
  • the millimeter wave data of the human body is processed into a millimeter wave grayscale image of the front or back of the human body.
  • the millimeter wave grayscale image of the human body refers to a millimeter wave grayscale image including only a human body contour region from which the human body clothing image and the background image are removed, or a millimeter wave grayscale image in which the human body contour region is placed on a solid color background. .
  • Step S120 Extract a foreign object region image in the millimeter wave grayscale image according to a preset foreign matter imaging characteristic.
  • the foreign matter may be a metal-based foreign matter such as a metal gun, a cutter, or a gold bullion, or may be a non-metallic foreign matter such as a chemical agent, an ivory, or a jade.
  • the preset foreign matter imaging characteristic specifically refers to:
  • a geometrically complex and well-defined geometrical area of the millimeter-wave grayscale image is determined as a non-metallic foreign matter.
  • the millimeter wave grayscale image is acquired based on a millimeter wave SAR (Synthetic Aperture Radar) imaging technology, and correspondingly, the preset foreign matter imaging characteristic specifically refers to:
  • the millimeter wave SAR images of different objects have different characteristics of speckle noise to identify foreign objects, thereby extracting the foreign object region image in the millimeter wave gray image.
  • step S120 specifically includes:
  • Step S121 Control a preset window to scan search on the millimeter wave grayscale image
  • Step S122 Extracting a window area image conforming to the preset foreign object imaging characteristic requirement as the foreign object area image.
  • the size and shape of the preset window may be set in advance according to actual needs, for example, according to the pixel size of the contour area of the human body in the acquired millimeter wave grayscale image.
  • a rectangular window having a pixel size of 80*80 is selected.
  • FIG. 3 it is an image of a foreign object area searched through a rectangular window.
  • step S121 specifically includes:
  • the control preset window slides the search at the preset step speed on the millimeter wave grayscale image.
  • the preset step speed can be set according to actual needs. In this embodiment, it is set to be 20 pixels per step.
  • the size of the preset window and the stepping speed determine its millimeter wave gray.
  • the scanning accuracy when sliding the search on the degree image the smaller the size of the preset window, the slower the stepping speed, the higher the scanning precision.
  • step S121 specifically includes:
  • the control preset window slides the search on the millimeter wave grayscale image at a preset step speed according to a preset search path.
  • the preset search path may be determined by the size and shape of the preset window and the size of the human contour region on the millimeter wave grayscale image. If the width of the preset window is greater than or equal to the maximum width of the contour region of the human body, You can slide straight to the bottom from the top of the contour area of the human body. If the length of the preset window is greater than or equal to the maximum length of the contour area of the human body, the right side can be slid from the left side of the contour area of the human body to the right side.
  • a rectangular window having a length and a width smaller than the length and width of the contour area of the human body is used, and the "hex" type is used from the left to the right, and the upper and lower fold line sliding search mode is used to search step by step.
  • step S122 the method further includes:
  • Step S123 If the window area image that meets the requirement of the preset foreign object imaging characteristic is not found, adjust the size or shape of the preset window;
  • Step S124 Control the window of the adjusted size or shape to scan the search on the millimeter wave grayscale image again.
  • the method may further include: After searching for the window area image that meets the requirements of the preset foreign object imaging characteristic, the preset step speed or the preset search path is adjusted.
  • Step S130 Calculate the foreign object region image according to a preset foreign object image recognition algorithm, and acquire a foreign matter image in the foreign object region image.
  • the foreign object area image may be classified into a background pixel area, a possible background pixel area, a target pixel area, and a possible target pixel area, and then each pixel pixel in the foreign object area image is calculated, and each Pixels are divided into different regions, and finally, through accurate classification and division calculation, it is determined whether the pixel points in the possible target pixel region belong to the target pixel region, and finally all the pixels belonging to the target pixel region are restored to digital images. , that is, get a foreign object image.
  • Step S140 Display the foreign matter image as a foreign matter detection result.
  • Fig. 4 shows a foreign matter image obtained after the calculation processing.
  • the millimeter wave grayscale image of the human body is acquired, and the foreign object region image in the millimeter wave grayscale image is extracted according to the preset foreign matter imaging characteristic, and the foreign object region image is calculated according to the preset foreign object image recognition algorithm, and the image is acquired.
  • the foreign matter image in the image of the foreign matter region can greatly improve the accuracy of foreign matter detection.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • this embodiment is a further refinement of step S130 in the first embodiment.
  • the preset foreign object image recognition algorithm in step S130 is an image segmentation algorithm.
  • Step S130 includes:
  • Step S131 classify all the pixel points in the foreign object area image.
  • all the pixel points in the foreign object area image may be divided into at least two categories according to the preset categorization model according to the RGB values of all the pixel points in the foreign object area image, and the preset classification model includes at least the preset target.
  • Regional model and preset background area model In a specific application, it may also be divided into more categories according to actual needs. For example, it may also include a possible background pixel area model and a possible target pixel area model, and the RGB value characteristics of the pixels belonging to different models are different.
  • the preset target area model is a model established according to the RGB value characteristics of the foreign object image.
  • Step S132 Acquire all pixel points belonging to the preset target area model, and establish a target pixel set
  • Step S133 Restore all the pixel points in the target pixel set to a digital image to obtain the foreign object image.
  • step S132 the method further includes:
  • the target pixel set is an empty set, the foreign object area image in the millimeter wave grayscale image is re-extracted according to the preset foreign object imaging characteristic.
  • the target pixel set is an empty set, that is, the pixel point that matches the RGB value characteristic of the foreign object image is not obtained, that is, the foreign object image is not acquired, so the foreign object area image in the millimeter wave gray image needs to be newly extracted. Then, the foreign matter image is calculated again by setting the foreign matter image recognition algorithm.
  • the image segmentation algorithm is specifically a clustering algorithm (K-means).
  • the step S130 may specifically include the following steps:
  • GMM Gaussian Mixture Model
  • the pixel points belonging to the target Gaussian model are acquired, and the pixel points belonging to the target Gaussian model are restored to a digital image, that is, the foreign object image is obtained.
  • the preset number may be set according to actual needs, and is preferably five in the embodiment, that is, a Gaussian mixture model including five Gaussian models is used to represent the RGB values of all the pixels in the foreign object region image. feature.
  • the calculation method for obtaining the pixel points belonging to the target Gaussian model is specifically:
  • the energy of the boundary pixel can reflect the degree of discontinuity between the pixel points m and n of the adjacent region. If the energy difference between two adjacent pixels is small, then they belong to the same target region or The possible performance of the same background area is very large. If the adjacent two pixels are very different, the possible performance of the same target area or the same background area is small.
  • the image of the foreign object region is clustered and segmented to obtain all the pixels belonging to the target Gaussian model.
  • the image of the foreign object region is segmented by the image segmentation algorithm, and the pixel points belonging to the foreign matter image in the foreign object region image are distinguished from the pixel points belonging to the background to obtain the pixel points belonging to the foreign matter image and restored to the foreign matter image.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • the millimeter wave image-based human body foreign body detecting system 100 provided in this embodiment is used to perform the method steps in the embodiment corresponding to FIG. 1.
  • the system 100 includes:
  • the human body grayscale image acquiring module 110 is configured to acquire a millimeter wave grayscale image of the human body
  • the foreign object region image extracting module 120 is configured to extract the foreign object region image in the millimeter wave grayscale image according to the preset foreign object imaging characteristic
  • the foreign object image calculation module 130 is configured to calculate the foreign object region image according to the preset foreign object image recognition algorithm, and acquire the foreign matter image in the foreign object region image;
  • the display module 140 is configured to display the foreign object image as a foreign matter detection result.
  • the human body gray image acquisition module 110 includes a millimeter wave transceiver and a millimeter wave imager; the foreign object region image extraction module 120 and the foreign matter image calculation module 130 may be image processing chips; and the display module 140 is a liquid crystal or LED display. display screen.
  • the foreign object region image extraction module 120 includes:
  • a window search unit configured to control a preset window to scan search on the millimeter wave grayscale image
  • An image extracting unit is configured to extract a window area image that meets a requirement of a predetermined foreign object imaging characteristic as the foreign object area image.
  • the foreign object region image extraction module 120 further includes:
  • a window adjusting unit configured to adjust a size or a shape of the preset window if a window area image that meets the requirement of the preset foreign object imaging characteristic is not searched;
  • the window search unit is further configured to control the window of the adjusted size or shape to scan the search on the millimeter wave grayscale image again.
  • the window search unit is specifically used to:
  • the control preset window slides the search at the preset step speed on the millimeter wave grayscale image.
  • the preset step speed can be set according to actual needs. In this embodiment, it is set to be 20 pixels per step.
  • the size of the preset window and the stepping speed determine the scanning precision of the sliding search on the millimeter wave gray image.
  • the window search unit is specifically used to:
  • the control preset window slides the search on the millimeter wave grayscale image at a preset step speed according to a preset search path.
  • the preset search path may be determined by the size and shape of the preset window and the size of the human contour region on the millimeter wave grayscale image. If the width of the preset window is greater than or equal to the maximum width of the contour region of the human body, You can slide straight to the bottom from the top of the contour area of the human body. If the length of the preset window is greater than or equal to the maximum length of the contour area of the human body, the right side can be slid from the left side of the contour area of the human body to the right side.
  • a rectangular window having a length and a width smaller than the length and width of the contour area of the human body is used, and the "hex" type is used from the left to the right, and the upper and lower fold line sliding search mode is used to search step by step.
  • the window adjusting unit is further configured to adjust a size or a shape of the preset window if a window area image that meets the preset foreign object imaging characteristic requirement is not searched;
  • the window search unit is further configured to control the window of the adjusted size or shape to scan the search on the millimeter wave grayscale image again.
  • the window adjusting unit is also used for After adjusting the window area image that meets the requirements of the preset foreign object imaging characteristic, the preset step speed or the preset search path is adjusted.
  • the millimeter wave grayscale image of the human body is acquired, and the foreign matter region image in the millimeter wave grayscale image is extracted according to the preset foreign matter imaging characteristic, and the image according to the preset foreign object image
  • the algorithm calculates the foreign object region image, and acquires the foreign matter image in the foreign object region image, which can greatly improve the accuracy of the foreign matter detection.
  • Embodiment 4 is a diagrammatic representation of Embodiment 4:
  • the foreign object image calculation module 130 provided in this embodiment is used to execute the method steps in the embodiment corresponding to FIG. 5.
  • the preset foreign object image recognition algorithm is an image segmentation algorithm
  • the foreign object image calculation module 130 includes:
  • a classification unit 131 configured to classify all pixel points in the foreign object area image
  • a target pixel set obtaining unit 132 configured to acquire all pixel points belonging to the preset target area model, and establish a target pixel set
  • the foreign object image obtaining unit 133 is configured to restore all the pixel points in the target pixel set to a digital image to obtain the foreign matter image.
  • the foreign object region image extraction module 120 is further configured to:
  • the target pixel set is an empty set, the foreign object area image in the millimeter wave grayscale image is re-extracted according to the preset foreign object imaging characteristic.
  • the target pixel set is an empty set, that is, the pixel point that matches the RGB value characteristic of the foreign object image is not obtained, that is, the foreign object image is not acquired, so the foreign object area image in the millimeter wave gray image needs to be newly extracted. Then, the foreign matter image is calculated again by setting the foreign matter image recognition algorithm.
  • the image segmentation algorithm is specifically a clustering algorithm (K-means).
  • the foreign object image computing module 130 may specifically include:
  • An RGB value acquiring unit configured to acquire RGB values of all pixels in the image of the foreign object region
  • a modeling unit configured to model the foreign object region image by using a Gaussian Mixture Model (GMM) including a preset number of Gaussian models;
  • GMM Gaussian Mixture Model
  • a determining unit configured to determine a Gaussian model corresponding to each pixel point according to the RGB values of all the pixel points;
  • the foreign matter image acquiring unit is configured to acquire pixel points belonging to the target Gaussian model according to the Gaussian model corresponding to each pixel point, and restore the pixel points belonging to the target Gaussian model to a digital image, that is, obtain the foreign object image.
  • the image of the foreign object region is segmented by the image segmentation algorithm, and the pixel points belonging to the foreign matter image in the foreign object region image are distinguished from the pixel points belonging to the background to obtain the pixel points belonging to the foreign matter image and restored to the foreign matter image.
  • the modules or units in all the embodiments of the present invention may be implemented by a general-purpose integrated circuit, such as a CPU (Central Processing Unit) or an ASIC (Application Specific Integrated Circuit).
  • a CPU Central Processing Unit
  • ASIC Application Specific Integrated Circuit
  • Modules or units in the system of the embodiments of the present invention may be combined, divided, and deleted according to actual needs, and each module or unit in the system functions in a one-to-one correspondence with the method steps, and the technical features in the system may be recorded in the method steps.
  • the technical characteristics are based on the increase.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geophysics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Geophysics And Detection Of Objects (AREA)
  • Image Processing (AREA)

Abstract

一种基于毫米波图像的人体异物检测方法，包括：获取人体的毫米波灰度图像（S110）；根据预设异物成像特性提取毫米波灰度图像中的异物区域图像（S120）；根据预设异物图像识别算法对异物区域图像进行计算，获取异物区域图像中的异物图像（S130）；显示异物图像，作为异物检测结果（S140）。还提供了一种基于毫米波图像的人体异物检测系统。

Description

一种基于毫米波图像的人体异物检测方法及系统 技术领域
本发明实施例属于毫米波成像技术领域，尤其涉及一种基于毫米波图像的人体异物检测方法及系统。
背景技术
毫米波在传播过程中具有一定的穿透性，通过毫米波成像技术能够安全方便的获得衣物遮挡下的金属或非金属类物品的图像，且不会对人体造成电离伤害，能够取代传统的X射线安检设备和金属探测器，应用于地铁、火车站、机场等公共场所的安全检查设备。
然而，现有的基于毫米波成像技术的安全检查设备，通常直接根据获得的人体的毫米波图像的灰度特性来检测人体中的异物，由于毫米波图像的灰度层次较低、清晰度低且具有明显的相干性斑状噪声，因此，直接利用人体的毫米波灰度图像的灰度特性来检测异物，效果并不理想。
发明内容
本发明实施例提供一种基于毫米波图像的人体异物检测方法及系统，旨在解决现有的基于毫米波成像技术的安全检查设备，直接根据获得的人体的毫米波图像的灰度特性来检测人体中的异物，效果并不理想的问题。
本发明实施例一方面提供一种基于毫米波图像的人体异物检测方法，所述方法包括：
获取人体的毫米波灰度图像；
根据预设异物成像特性提取所述毫米波灰度图像中的异物区域图像；
根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像；
显示所述异物图像，作为异物检测结果。
本发明另一方面还一种基于毫米波图像的人体异物检测系统，所述系统包括：
人体灰度图像获取模块，用于获取人体的毫米波灰度图像；
异物区域图像提取模块，用于根据预设异物成像特性提取所述毫米波灰度图像中的异物区域图像；
异物图像计算模块，用于根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像；
显示模块，用于显示所述异物图像，作为异物检测结果。
本发明实施例通过获取人体的毫米波灰度图像，并根据预设异物成像特性提取毫米波灰度图像中的异物区域图像，根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像，可大大提高异物检测的准确性。
附图说明
图1是本发明实施例一提供的基于毫米波图像的人体异物检测 方法的基本流程框图；
图2是本发明实施例一提供的毫米波灰度图像的示意图；
图3是本发明实施例一提供的异物区域图像的示意图；
图4是本发明实施例一提供的异物图像的示意图；
图5是本发明实施例二提供的步骤S130的流程框图；
图6是本发明实施例三提供的基于毫米波图像的人体异物检测系统的基本结构框图；
图7是本发明实施例四提供的异物图像计算模块的结构框图。
具体实施方式
为了使本技术领域的人员更好地理解本发明方案，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地描述，显然，所描述的实施例是本发明一部分的实施例，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，都应当属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语“包括”以及它们任何变形，意图在于覆盖不排他的包含。例如包含一系列步骤或单元的过程、方法或系统、产品或设备没有限定于已列出的步骤、模块或单元，而是可选地还包括没有列出的步骤、模块或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤、模块或单元。
实施例一：
如图1所示，本实施例所提供的基于毫米波图像的人体异物检测方法，其包括：
步骤S110：获取人体的毫米波灰度图像。
在具体应用中，可以使人体双手上扬举过头顶或双手平举至与肩部同高的位置站立，也可以使人体双手自然下垂站立，或者采用其他符合安检标准的站姿均可，本发明实施例不对人体的站姿作特别限定，然后利用毫米波数据采集设备(例如，毫米波收发机)获取人体正面或背面的毫米波数据，并利用毫米波成像系统(例如，毫米波成像仪)将人体的毫米波数据处理为人体正面或背面的毫米波灰度图像。
本实施例中，人体的毫米波灰度图像是指去除了人体衣物图像和背景图像的仅包含人体轮廓区域的毫米波灰度图像或者将人体轮廓区域放置在纯色背景上的毫米波灰度图像。
如图2所示为成年男性双手上扬高举过头顶拍摄得到的人体的毫米波灰度图像。
步骤S120：根据预设异物成像特性提取所述毫米波灰度图像中的异物区域图像。
在具体应用中，所述异物可以是金属类的枪支、刀具、金块等金属类异物，也可以是化学药剂、象牙、玉石等非金属类异物。
在一实施例中，预设异物成像特性具体是指：
将所述毫米波灰度图像中灰度值大于预设灰度阈值且轮廓分明的区域，确定为金属类异物；
将所述毫米波灰度图像中纹理复杂、轮廓分明的几何图形区域，确定为非金属类异物。
在一实施例中，所述毫米波灰度图像基于毫米波SAR(合成孔径雷达，Synthetic Aperture Radar)成像技术获取，对应的，预设异物成像特性具体是指：
根据毫米波SAR成像技术的成像特点，即不同物体的毫米波SAR图像具有不同的相干斑噪声的特点来识别异物，从而在毫米波灰度图像中提取出异物区域图像。
在一实施例中，步骤S120具体包括：
步骤S121：控制预设窗口在所述毫米波灰度图像上滑动搜索；
步骤S122：提取符合预设异物成像特性要求的窗口区域图像，作为所述异物区域图像。
在具体应用中，预设窗口的大小和形状可以根据实际需要事先设定，例如，根据获取的毫米波灰度图像中人体轮廓区域的像素大小来确定。本实施例中，选择像素大小为80*80的矩形窗口。
如图3所示，为通过一矩形窗口搜索到的异物区域图像。
在具体应用中，步骤S121具体包括：
控制预设窗口在所述毫米波灰度图像上以预设步进速度滑动搜索。
在具体应用中，预设步进速度可以根据实际需要进行设定，本实施例中，设定为每移动一下窗口步进20个像素单位。
在具体应用中，预设窗口的大小和步进速度决定了其在毫米波灰 度图像上滑动搜索时的扫描精度，预设窗口的尺寸越小，步进速度越慢，则扫描精度越高。
在具体应用中，步骤S121具体还包括：
控制预设窗口在所述毫米波灰度图像上以预设步进速度按照预设搜索路径滑动搜索。
在具体应用中，预设搜索路径可以由预设窗口的大小、形状和毫米波灰度图像上人体轮廓区域的大小来决定，若预设窗口的宽度大于或等于人体轮廓区域的最大宽度，则可以从人体轮廓区域的顶部直线滑动搜索到底部。若预设窗口的长度大于或等于人体轮廓区的最大长度，则可以从人体轮廓区域的左侧直线滑动搜索到右侧。本实施例中，采用长、宽均小于人体轮廓区域的长度和宽度的矩形窗口，采用“己”字型从左到右，由上而下的折线滑动搜索方式，逐步搜索。
在一实施例中，步骤S122之后还包括：
步骤S123：若未搜索到符合所述预设异物成像特性要求的窗口区域图像，则调整所述预设窗口的大小或形状；
步骤S124：控制所述调整了大小或形状的窗口重新在所述毫米波灰度图像上滑动搜索。
在具体应用中，未搜索到符合所述预设异物成像特性要求的窗口区域图像，还有可能是因为搜索路径不理想或者是步进速度过快，因此，步骤S122之后还可以包括：若未搜索到符合所述预设异物成像特性要求的窗口区域图像，则调整所述预设步进速度或预设搜索路径。
步骤S130：根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像。
在具体应用中，可以将异物区域图像分类为背景像素区域、可能的背景像素区域、目标像素区域和可能的目标像素区域，然后通过对异物区域图像中所有像素点的像素特性进行计算，将各像素点划分到不同的区域，最后通过精确的归类划分计算，确定可能的目标像素区域中的像素点是否属于目标像素区域，最后将获得的所有属于目标像素区域的所有像素点还原为数字图像，即得到异物图像。
步骤S140：显示所述异物图像，作为异物检测结果。
在具体应用中，显示异物图像时，仅显示窗口中的目标像素区域，背景像素区域替换为纯色背景。
图4所示为经过计算处理之后得到的异物图像。
本实施例通过获取人体的毫米波灰度图像，并根据预设异物成像特性提取毫米波灰度图像中的异物区域图像，根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像，可大大提高异物检测的准确性。
实施例二：
如图5所示，本实施例是对实施例一中步骤S130的进一步细化。
在具体应用中，步骤S130中的预设异物图像识别算法为图像分割算法。
步骤S130包括：
步骤S131：对所述异物区域图像中的所有像素点进行分类。
在具体应用中，可以根据异物区域图像中的所有像素点的RGB值，按照预设分类模型将异物区域图像中的所有像素点分为至少两个类别，该预设分类模型至少包括预设目标区域模型和预设背景区域模型。在具体应用中，还可以根据实际需要分为更多类别，例如，还可以包括可能的背景像素区域模型和可能的目标像素区域模型，属于不同模型的像素点的RGB值特性不同。
本实施例中，预设目标区域模型即是按照异物图像的RGB值特点建立的模型。
步骤S132：获取属于预设目标区域模型的所有像素点，建立目标像素集合；
步骤S133：将所述目标像素集合中的所有像素点还原为数字图像，得到所述异物图像。
在一实施例中，步骤S132之后还包括：
若所述目标像素集合为空集，则根据预设异物成像特性重新提取所述毫米波灰度图像中的异物区域图像。
在具体应用中，目标像素集合为空集即表明没有获取到符合异物图像的RGB值特点的像素点，即没有获取到异物图像，因此需要重新提取所述毫米波灰度图像中的异物区域图像，然后再次通过设异物图像识别算法计算得到异物图像。
本实施例中，所述图像分割算法具体为聚类算法(K-means)，对应的，步骤S130具体可以包括如下步骤：
获取所述异物区域图像中所有像素点的RGB值；
采用包括预设个数的高斯模型的高斯混合模型(Gaussian Mixture Model，GMM)对所述异物区域图像进行拟合建模；
根据所述所有像素点的RGB值，判断每个像素点所对应的高斯模型；
根据每个像素点所对应的高斯模型，获取属于目标高斯模型的像素点，并将属于目标高斯模型的像素点还原为数字图像，即得到所述异物图像。
在具体应用中，所述预设个数可根据实际需要设定，本实施例中优选为5个，即采用包括5个高斯模型的高斯混合模型来表示异物区域图像中所有像素点的RGB值特征。
在具体应用中，获取属于目标高斯模型的像素点的计算方法具体为：
设定整个异物区域图像的吉布斯(Gibbs)能量为E(a,k,θ,z)＝U(a,k,θ,z)+V(a,z),；
设定目标区域U中的所有的像素点的能量为
Figure PCTCN2017096101-appb-000001
设定目标区域U的边界像素点的能量
Figure PCTCN2017096101-appb-000002
则所要计算得到的属于目标高斯模型的像素点的能量为
Figure PCTCN2017096101-appb-000003
其中，θ＝{π(a,k),μ(a,k),∑(a,k),a＝0,1,k＝1...K}。
在具体应用中，β与异物区域图像本身的对比度有关，本实施例中优选参数γ＝40。
在具体应用中，边界像素点的能量能够体现相邻区域的像素点m和n之间的不连续的程度，如果两相邻像素点之间的能量差别很小，那么它们属于同一目标区域或者同一背景区域的可能性能很大，如果相邻两像素差别很大，那么它们属于同一目标区域或者同一背景区域的可能性能很小，本实施例中，即根据相邻的边界像素点之间的能量差别特点，对异物区域图像进行聚类分割，以获取属于目标高斯模型的所有像素点。
本实施例通过图像分割算法对异物区域图像进行分割，可对异物区域图像中属于异物图像的像素点与属于背景的像素点进行区分，以获得属于异物图像的像素点并还原为异物图像。
实施例三：
如图6所示，本实施例提供的基于毫米波图像的人体异物检测系统100用于执行图1所对应的实施例中的方法步骤，该系统100包括：
人体灰度图像获取模块110，用于获取人体的毫米波灰度图像；
异物区域图像提取模块120，用于根据预设异物成像特性提取所述毫米波灰度图像中的异物区域图像；
异物图像计算模块130，用于根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像；
显示模块140，用于显示所述异物图像，作为异物检测结果。
在具体应用中，人体灰度图像获取模块110包括毫米波收发机和毫米波成像仪；异物区域图像提取模块120和异物图像计算模块130可以为图像处理芯片；显示模块140为液晶或LED显示器等显示设备。
在一实施例中，异物区域图像提取模块120包括：
窗口搜索单元，用于控制预设窗口在所述毫米波灰度图像上滑动搜索；
图像提取单元，用于提取符合预设异物成像特性要求的窗口区域图像，作为所述异物区域图像。
在另一实施例中，异物区域图像提取模块120还包括：
窗口调整单元，用于若未搜索到符合所述预设异物成像特性要求的窗口区域图像，则调整所述预设窗口的大小或形状；
所述窗口搜索单元还用于控制所述调整了大小或形状的窗口重新在所述毫米波灰度图像上滑动搜索。
在具体应用中，窗口搜索单元，具体用于：
控制预设窗口在所述毫米波灰度图像上以预设步进速度滑动搜索。
在具体应用中，预设步进速度可以根据实际需要进行设定，本实施例中，设定为每移动一下窗口步进20个像素单位。
在具体应用中，预设窗口的大小和步进速度决定了其在毫米波灰度图像上滑动搜索时的扫描精度，预设窗口的尺寸越小，步进速度越慢，则扫描精度越高。
在具体应用中，窗口搜索单元，具体还用于：
控制预设窗口在所述毫米波灰度图像上以预设步进速度按照预设搜索路径滑动搜索。
在具体应用中，预设搜索路径可以由预设窗口的大小、形状和毫米波灰度图像上人体轮廓区域的大小来决定，若预设窗口的宽度大于或等于人体轮廓区域的最大宽度，则可以从人体轮廓区域的顶部直线滑动搜索到底部。若预设窗口的长度大于或等于人体轮廓区的最大长度，则可以从人体轮廓区域的左侧直线滑动搜索到右侧。本实施例中，采用长、宽均小于人体轮廓区域的长度和宽度的矩形窗口，采用“己”字型从左到右，由上而下的折线滑动搜索方式，逐步搜索。
在一实施例中，窗口调整单元，还用于若未搜索到符合所述预设异物成像特性要求的窗口区域图像，则调整所述预设窗口的大小或形状；
窗口搜索单元，还用于控制所述调整了大小或形状的窗口重新在所述毫米波灰度图像上滑动搜索。
在具体应用中，未搜索到符合所述预设异物成像特性要求的窗口区域图像，还有可能是因为搜索路径不理想或者是步进速度过快，因此，窗口调整单元，还用于可以用于：若未搜索到符合所述预设异物成像特性要求的窗口区域图像，则调整所述预设步进速度或预设搜索路径。
本实施例通过获取人体的毫米波灰度图像，并根据预设异物成像特性提取毫米波灰度图像中的异物区域图像，根据预设异物图像识 别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像，可大大提高异物检测的准确性。
实施例四：
如图7所示，本实施例提供的异物图像计算模块130用于执行图5所对应的实施例中的方法步骤。
本实施例中，所述预设异物图像识别算法为图像分割算法；
对应的，异物图像计算模块130包括：
分类单元131，用于对所述异物区域图像中的所有像素点进行分类；
目标像素集合获取单元132，用于获取属于预设目标区域模型的所有像素点，建立目标像素集合；
异物图像获取单元133，用于将所述目标像素集合中的所有像素点还原为数字图像，得到所述异物图像。
在一实施例中，异物区域图像提取模块120，还用于：
若所述目标像素集合为空集，则根据预设异物成像特性重新提取所述毫米波灰度图像中的异物区域图像。
在具体应用中，目标像素集合为空集即表明没有获取到符合异物图像的RGB值特点的像素点，即没有获取到异物图像，因此需要重新提取所述毫米波灰度图像中的异物区域图像，然后再次通过设异物图像识别算法计算得到异物图像。
本实施例中，所述图像分割算法具体为聚类算法(K-means)，对应的，异物图像计算模块130具体可以包括：
RGB值获取单元，用于获取所述异物区域图像中所有像素点的RGB值；
建模单元，用于采用包括预设个数的高斯模型的高斯混合模型(Gaussian Mixture Model，GMM)对所述异物区域图像进行拟合建模；
判断单元，用于根据所述所有像素点的RGB值，判断每个像素点所对应的高斯模型；
异物图像获取单元，用于根据每个像素点所对应的高斯模型，获取属于目标高斯模型的像素点，并将属于目标高斯模型的像素点还原为数字图像，即得到所述异物图像。
本实施例通过图像分割算法对异物区域图像进行分割，可对异物区域图像中属于异物图像的像素点与属于背景的像素点进行区分，以获得属于异物图像的像素点并还原为异物图像。
本发明所有实施例中的模块或单元，可以通过通用集成电路，例如CPU(Central Processing Unit，中央处理器)，或通过ASIC(Application Specific Integrated Circuit，专用集成电路)来实现。
本发明实施例方法中的步骤可以根据实际需要进行顺序调整、合并和删减。
本发明实施例系统中的模块或单元可以根据实际需要进行合并、划分和删减，且系统中各模块或单元的作用于方法步骤一一对应，系统中的技术特征均可以以方法步骤所记载的技术特征为依据进行增加。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (10)

  1. 一种基于毫米波图像的人体异物检测方法，其特征在于，所述方法包括：
    获取人体的毫米波灰度图像；
    根据预设异物成像特性提取所述毫米波灰度图像中的异物区域图像；
    根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像；
    显示所述异物图像，作为异物检测结果。
  2. 如权利要求1所述的基于毫米波图像的人体异物检测方法，其特征在于，所述获取所述毫米波灰度中的异物区域图像，包括：
    控制预设窗口在所述毫米波灰度图像上滑动搜索；
    提取符合预设异物成像特性要求的窗口区域图像，作为所述异物区域图像。
  3. 如权利要求2所述的基于毫米波图像的人体异物检测方法，其特征在于，所述控制预设窗口在所述毫米波灰度图像上滑动搜索之后，还包括：
    若未搜索到符合所述预设异物成像特性要求的窗口区域图像，则调整所述预设窗口的大小或形状；
    控制所述调整了大小或形状的窗口重新在所述毫米波灰度图像上滑动搜索。
  4. 如权利要求1所述的基于毫米波图像的人体异物检测方法，其特征在于，
    所述预设异物图像识别算法为图像分割算法；
    对应的，所述根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像，包括：
    对所述异物区域图像中的所有像素点进行分类；
    获取属于预设目标区域模型的所有像素点，建立目标像素集合；
    将所述目标像素集合中的所有像素点还原为数字图像，得到所述异物图像。
  5. 如权利要求4所述的基于毫米波图像的人体异物检测方法，其特征在于，所述获取属于所述预设目标区域模型的所有像素点，建立目标像素集合之后，还包括：
    若所述目标像素集合为空集，则根据预设异物成像特性重新提取所述毫米波灰度图像中的异物区域图像。
  6. 一种基于毫米波图像的人体异物检测系统，其特征在于，所述系统包括：
    人体灰度图像获取模块，用于获取人体的毫米波灰度图像；
    异物区域图像提取模块，用于根据预设异物成像特性提取所述毫米波灰度图像中的异物区域图像；
    异物图像计算模块，用于根据预设异物图像识别算法对所述异物区域图像进行计算，获取所述异物区域图像中的异物图像；
    显示模块，用于显示所述异物图像，作为异物检测结果。
  7. 如权利要求6所述的基于毫米波图像的人体异物检测系统，其特征在于，所述异物区域图像提取模块包括：
    窗口搜索单元，用于控制预设窗口在所述毫米波灰度图像上滑动搜索；
    图像提取单元，用于提取符合预设异物成像特性要求的窗口区域图像，作为所述异物区域图像。
  8. 如权利要求7所述的基于毫米波图像的人体异物检测系统，其特征在于，所述异物区域图像提取模块还包括：
    窗口调整单元，用于若未搜索到符合所述预设异物成像特性要求的窗口区域图像，则调整所述预设窗口的大小或形状；
    所述窗口搜索单元还用于控制所述调整了大小或形状的窗口重新在所述毫米波灰度图像上滑动搜索。
  9. 如权利要求6所述的基于毫米波图像的人体异物检测系统，其特征在于，
    所述预设异物图像识别算法为图像分割算法；
    对应的，所述异物图像计算模块包括：
    分类单元，用于对所述异物区域图像中的所有像素点进行分类；
    目标像素集合获取单元，用于获取属于预设目标区域模型的所有像素点，建立目标像素集合；
    异物图像获取单元，用于将所述目标像素集合中的所有像素点还原为数字图像，得到所述异物图像。
  10. 如权利要求9所述的基于毫米波图像的人体异物检测系统， 其特征在于，所述异物区域图像提取模块，还用于：
    若所述目标像素集合为空集，则根据预设异物成像特性重新提取所述毫米波灰度图像中的异物区域图像。
PCT/CN2017/096101 2016-09-27 2017-08-04 一种基于毫米波图像的人体异物检测方法及系统 WO2018059125A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP17854578.6A EP3521864A4 (en) 2016-09-27 2017-08-04 METHOD AND SYSTEM FOR DETECTING FOREIGN MATTERS OF THE HUMAN BODY ON A MILLIMETER WAVE IMAGE BASE
US16/336,772 US11194073B2 (en) 2016-09-27 2017-08-04 Millimeter wave image based human body foreign object detection method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610855267.8A CN106371148B (zh) 2016-09-27 2016-09-27 一种基于毫米波图像的人体异物检测方法及系统
CN201610855267.8 2016-09-27

Publications (1)

Publication Number Publication Date
WO2018059125A1 true WO2018059125A1 (zh) 2018-04-05

Family

ID=57897146

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/096101 WO2018059125A1 (zh) 2016-09-27 2017-08-04 一种基于毫米波图像的人体异物检测方法及系统

Country Status (4)

Country Link
US (1) US11194073B2 (zh)
EP (1) EP3521864A4 (zh)
CN (1) CN106371148B (zh)
WO (1) WO2018059125A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111539441A (zh) * 2019-01-30 2020-08-14 杭州芯影科技有限公司 一种基于毫米波安检图像的隐匿物检测方法及系统

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106371148B (zh) * 2016-09-27 2019-05-03 华讯方舟科技有限公司 一种基于毫米波图像的人体异物检测方法及系统
CN107801368B (zh) * 2017-08-08 2019-01-15 朗丝窗饰有限公司 避免辐射对人体身心健康造成伤害的系统
CN210155349U (zh) * 2018-03-09 2020-03-17 同方威视技术股份有限公司 可扩展式毫米波安检系统及扫描单元
CN108985196A (zh) * 2018-06-29 2018-12-11 深圳市华讯方舟太赫兹科技有限公司 评价安检仪准确度的方法、终端和具有存储功能的装置
CN109543582A (zh) * 2018-11-15 2019-03-29 杭州芯影科技有限公司 基于毫米波图像的人体异物检测方法
CN109711331B (zh) * 2018-12-25 2022-08-09 山东雷诚电子科技有限公司 一种毫米波雷达安检仪异物检测方法
CN112764120A (zh) * 2020-12-29 2021-05-07 深圳市创艺龙电子科技有限公司 金属探测成像系统和具有其的扫描设备
CN115311684B (zh) * 2022-08-05 2023-03-28 杭州电子科技大学 一种毫米波图像多角度检出结果的整合方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1183996A (ja) * 1997-09-03 1999-03-26 Omron Corp ミリ波検出装置
JP2006258496A (ja) * 2005-03-15 2006-09-28 Fujitsu Ltd ミリ波撮像装置
KR20090092660A (ko) * 2008-02-27 2009-09-01 동국대학교 산학협력단 Ccd 영상으로 보정하는 나노구조 반도체 소자를 이용한 수동 밀리미터파 은닉 물체 감지 장치
US8116520B2 (en) * 2007-06-20 2012-02-14 Microsemi Corporation System and method for overlaying computer generated highlights in a display of millimeter wave imagery
CN104459822A (zh) * 2014-12-05 2015-03-25 同方威视技术股份有限公司 人体安全检查设备
CN105513035A (zh) * 2014-09-22 2016-04-20 北京计算机技术及应用研究所 一种被动毫米波图像中人体隐匿物品的检测方法及其系统
CN106371148A (zh) * 2016-09-27 2017-02-01 华讯方舟科技有限公司 一种基于毫米波图像的人体异物检测方法及系统

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100406113B1 (ko) * 2000-11-27 2003-11-15 학교법인 동국학원 엑스선 신호와 밀리미터파 신호를 이용한 검색장치
US7365672B2 (en) * 2001-03-16 2008-04-29 Battelle Memorial Institute Detection of a concealed object
WO2005086620A2 (en) * 2003-10-10 2005-09-22 L-3 Communications Security And Detection Systems Mmw contraband screening system
US7123185B2 (en) * 2004-04-14 2006-10-17 Safeview, Inc. Enhanced surveilled subject imaging
US7386150B2 (en) * 2004-11-12 2008-06-10 Safeview, Inc. Active subject imaging with body identification
DE102006049152A1 (de) * 2006-10-18 2008-04-24 Ludwig-Maximilians-Universität München Anordnung und Verfahren zum Detektieren eines an einem Körper angeordneten Gegenstands, insbesondere zur Durchführung einer Sicherheitskontrolle
US7804442B2 (en) * 2007-01-24 2010-09-28 Reveal Imaging, Llc Millimeter wave (MMW) screening portal systems, devices and methods
JP2010008272A (ja) * 2008-06-27 2010-01-14 Maspro Denkoh Corp ミリ波撮像装置
JP2010008273A (ja) * 2008-06-27 2010-01-14 Maspro Denkoh Corp ミリ波撮像装置
WO2010001377A2 (en) * 2008-07-01 2010-01-07 Smiths Detection Ireland Limited Identification of potential threat materials using active electromagnetic waves
CN101887518B (zh) * 2010-06-17 2012-10-31 北京交通大学 人体检测装置与方法
CN102567733B (zh) * 2011-12-30 2014-05-07 北京华航无线电测量研究所 一种微波图像非合作情况下人体隐私部位检测与遮挡方法
US8938114B2 (en) * 2012-01-11 2015-01-20 Sony Corporation Imaging device and method for imaging hidden objects
CN102697503B (zh) * 2012-02-29 2015-03-18 北京无线电计量测试研究所 一种基于毫米波成像的人体检测方法
CN102629315B (zh) * 2012-02-29 2016-07-06 北京无线电计量测试研究所 一种隐匿物品的自动检测和识别装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1183996A (ja) * 1997-09-03 1999-03-26 Omron Corp ミリ波検出装置
JP2006258496A (ja) * 2005-03-15 2006-09-28 Fujitsu Ltd ミリ波撮像装置
US8116520B2 (en) * 2007-06-20 2012-02-14 Microsemi Corporation System and method for overlaying computer generated highlights in a display of millimeter wave imagery
KR20090092660A (ko) * 2008-02-27 2009-09-01 동국대학교 산학협력단 Ccd 영상으로 보정하는 나노구조 반도체 소자를 이용한 수동 밀리미터파 은닉 물체 감지 장치
CN105513035A (zh) * 2014-09-22 2016-04-20 北京计算机技术及应用研究所 一种被动毫米波图像中人体隐匿物品的检测方法及其系统
CN104459822A (zh) * 2014-12-05 2015-03-25 同方威视技术股份有限公司 人体安全检查设备
CN106371148A (zh) * 2016-09-27 2017-02-01 华讯方舟科技有限公司 一种基于毫米波图像的人体异物检测方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3521864A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111539441A (zh) * 2019-01-30 2020-08-14 杭州芯影科技有限公司 一种基于毫米波安检图像的隐匿物检测方法及系统

Also Published As

Publication number Publication date
US11194073B2 (en) 2021-12-07
EP3521864A1 (en) 2019-08-07
CN106371148B (zh) 2019-05-03
CN106371148A (zh) 2017-02-01
US20200333504A1 (en) 2020-10-22
EP3521864A4 (en) 2020-06-03

Similar Documents

Publication Publication Date Title
WO2018059125A1 (zh) 一种基于毫米波图像的人体异物检测方法及系统
US8611604B2 (en) Object detection device
US10074028B2 (en) Real-time diet assessment and food identification using thermal imaging
JP6719457B2 (ja) 画像の主要被写体を抽出する方法とシステム
US9639748B2 (en) Method for detecting persons using 1D depths and 2D texture
Ikemura et al. Real-time human detection using relational depth similarity features
JP6125188B2 (ja) 映像処理方法及び装置
CN104573614B (zh) 用于跟踪人脸的设备和方法
WO2017190656A1 (zh) 行人再识别方法和装置
US8620024B2 (en) System and method for dynamic gesture recognition using geometric classification
He et al. A method of green litchi recognition in natural environment based on improved LDA classifier
US20160335491A1 (en) Method and device for detecting gathering of objects based on stereo vision as well as non-transitory computer-readable medium
US20140177915A1 (en) Method and apparatus for detecting object
JP2017033469A (ja) 画像識別方法、画像識別装置及びプログラム
WO2019064375A1 (ja) 情報処理装置、制御方法、及びプログラム
CN108961158B (zh) 一种图像合成方法及装置
CN103810478A (zh) 一种坐姿检测方法和装置
TWI415032B (zh) Object tracking method
EP3506148A1 (en) Millimetre wave image-based human body foreign matter detection method and system
US20190138802A1 (en) Method and apparatus for detecting human body gender in microwave image
Chansri et al. Reliability and accuracy of Thai sign language recognition with Kinect sensor
JP2011165170A (ja) 対象物検出装置及びプログラム
CN107704864B (zh) 基于图像对象性语义检测的显著目标检测方法
CN104517300A (zh) 基于颜色统计特征的视觉判别跟踪方法
Saha et al. An approach to detect the region of interest of expressive face images

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: 17854578

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2017854578

Country of ref document: EP

Effective date: 20190429