CN103777252B - Hazardous material detection method - Google Patents
Hazardous material detection method Download PDFInfo
- Publication number
- CN103777252B CN103777252B CN201410062083.7A CN201410062083A CN103777252B CN 103777252 B CN103777252 B CN 103777252B CN 201410062083 A CN201410062083 A CN 201410062083A CN 103777252 B CN103777252 B CN 103777252B
- Authority
- CN
- China
- Prior art keywords
- image
- rgb format
- infrared
- raw rgb
- infrared image
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a kind of hazardous material detection device and detection method, described hazardous material detection device includes visible light image sensor, infrared image sensor, processor, video memory and display；This hazardous material detection device is after visible light image sensor and infrared image sensor gather the visible images of raw RGB format and the infrared image of raw RGB format, image is processed and shows by described processor, and personnel's normal pass, detection i.e. can be realized without stopping, thus improve detection efficiency, bigger Pedestrian flow detection, large range of detection can be met and multiple target is detected simultaneously.
Description
Technical field
The present invention relates to a kind of article detection device and detection method, particularly relate to a kind of hazardous material detection device and detection method.
Background technology
Current all kinds of insurgent violence event emerges in an endless stream, and cutter injury case and public place violence attack of terrorism case etc. are held in such as campus, and these cases seriously threaten the stable of society and the security of the lives and property of the people.In order to avoid occurring again of this kind of case, country has put into substantial amounts of human and material resources and financial resources in terms of security, but has produced little effect.
The dangerous article detection system used at present is formula detecting system (stand-by) of stopping, such as metal detector and X ray tester etc., this type of system typically requires Security Personnel and detects a suspect one by one, when flow of the people is excessive, or need detection in a big way, or when detecting multiple target, such system, owing to detecting the inefficiency of dangerous material, just cannot meet above-mentioned requirements simultaneously.
Visible images has high-resolution feature, but cannot embody the temperature information of object, the most just cannot be individually used for the detection of dangerous material.Although the resolution ratio of infrared image is relatively low, but the image-forming principle of infrared image is temperature difference based on object, the temperature information of object can be fully demonstrated, and owing to gun, controlled knife etc. have the metallic article of menace, i.e. dangerous material are generally and the background object such as human body has a bigger temperature difference, all can be detected this kind of dangerous material being hidden in human body or other objects by infrared image sensor.Additionally, the high-resolution advantage of visible images can also be utilized, by information fusion such as the shape of dangerous material, size and locations in visible images, and do not change the color information of original visible images, thus fundamentally change the traditional method of the single-sensor that dangerous material information can only be shown in x-ray image, infrared image or the microwave and millimeter wave image of low resolution.
YIQ (luminance, chrominance) color space form, its representative image brightness (luminance) is independent with the part of representative image color (chrominance) information；It is to say, can't the color information of the original image of this change when the monochrome information of image individually being extracted and it being processed；Vice versa.Just because of this feature of YIQ color space form, it is possibly realized for solving effectively to detect the technical problem of dangerous material.
Summary of the invention
First purpose of the present invention is to provide a kind of hazardous material detection device that can effectively detect dangerous goods.Second object of the present invention is to provide a kind of hazardous material detection method that can effectively detect dangerous goods.
The present invention solves first technical problem and adopts the following technical scheme that a kind of hazardous material detection device, including visible light image sensor, infrared image sensor, processor, video memory and display；
Described visible light image sensor signal is connected to described processor, to gather the visible images of the detected object comprising described dangerous material, and by the transmission of described visible images to described processor；
Described infrared image sensor signal is connected to described processor, to gather the infrared image of the detected object comprising described dangerous material, and by described infrared image delivery to described processor；
Described visible images and infrared image are processed by described processor, the visible images after being processed and the infrared image after process；
Described video memory signal is connected to described processor, with the infrared image after storing the visible images after described process and processing；
Described monitor signal is connected to described processor, to show the image after described process.
Optionally, described video memory and/or display are connected to described processor by WIFI wireless data transfer module, Bluetooth wireless data transport module or Zigbee wireless data transfer module.
The present invention solves second technical problem and adopts the following technical scheme that a kind of method utilizing described hazardous material detection device detection dangerous material, comprises the following steps:
S10, utilize described visible light image sensor and infrared image sensor that the detected object comprising described dangerous material is carried out IMAQ, respectively obtain the visible images of raw RGB format and the infrared image of raw RGB format；
S20, the visible images of raw RGB format is converted to the visible images of YIQ form；
S30, generate visible ray gray level image according to the visible images of described YIQ form；
S40, the infrared image of raw RGB format is converted into infrared hybrid optical system；
S50, described infrared hybrid optical system is carried out pixel value reversion, the infrared hybrid optical system after being inverted；
S60, the infrared hybrid optical system after described visible ray gray level image, infrared hybrid optical system and reversion is carried out image co-registration, the image after being merged；
S70, by after described fusion image generate YIQ form fused images；
S80, the fused images of described YIQ form is converted to the fused images of rgb format；
S90, the fused images of described rgb format is shown on the display.
Optionally, after described step S10, before step S20, also include step S15:
S15, the visible images of raw RGB format and the infrared image of raw RGB format are carried out calibration process and frame synchronization process so that the pixel of the visible images of described raw RGB format and the infrared image of described raw RGB format and content matching.
Optionally, described step S80 and S90 are replaced with:
S100, the infrared image of raw RGB format is converted into the infrared image of YIQ form；
S110, the fused images of YIQ form and the infrared image of YIQ form are carried out image co-registration, obtain final fused images；
S120, described final fused images is converted into final gray level image；
S130, described final gray level image is carried out binary conversion treatment, obtain binaryzation gray level image；
S140, by the visual image fusion of the white portion in described binaryzation gray level image Yu described raw RGB format, and obtain image to be output；
S150, product information the most dangerous in described binaryzation gray level image is judged；
S160, when, in step S150, not finding dangerous material, the visible images of raw RGB format is shown on the display；
S170, when in step S150, find dangerous material, described image to be output shown on the display.
Optionally, described step S130 specifically includes following steps:
S1301, setting threshold value；
S1302, will be greater than being set to 1 equal to the value of the pixel of described threshold value；
S1303, the value of the pixel less than described threshold value is set to 0 obtains binaryzation gray level image.
Optionally, described step S140 specifically includes following steps:
S1401, the image after binaryzation is carried out three-dimensional expansion；
S1402, the pixel that value is 1 of described pixel is replaced the pixel of the visible images of the raw RGB format of same position, and the value of the pixel of the visible images of the raw RGB format of described same position is set to (1,1,1).
Optionally, the algorithm of the image co-registration in described step S60 is: linearity blending algorithm or Image Fusion based on wavelet transform.
There is advantages that
nullThe hazardous material detection device of the present invention includes visible light image sensor、Infrared image sensor and processor，The visible images of the raw RGB format that described visible light image sensor is gathered is transferred to processor，The infrared image delivery of the raw RGB format that described infrared image sensor is gathered is to processor，The visible images of described raw RGB format and the infrared image of raw RGB format are processed and merge by described processor，And the image after exporting described process by described display and merge，With the infrared image of the definition and raw RGB format that make full use of the visible images of the described raw RGB format sensitiveness to the temperature of described dangerous material，Realize detection (the such as gun to described dangerous material，Controlled knife)，This hazardous material detection device is after the infrared image of the visible images and raw RGB format that gather raw RGB format，Personnel's normal pass，Detection i.e. can be realized without stopping，Thus improve detection efficiency，Bigger Pedestrian flow detection can be met、Large range of detection and simultaneously multiple target being detected.
nullThe hazardous material detection method of the present invention is owing to have employed above-mentioned hazardous material detection device，And the visible images of the raw RGB format by being gathered by described visible light image sensor is converted into the visible images of YIQ form，The infrared image of the raw RGB format gathered by described infrared image sensor is converted into the infrared hybrid optical system after infrared hybrid optical system and reversion，Visible images to described YIQ form、Infrared hybrid optical system after described infrared hybrid optical system and described reversion merges，Image after merging is converted into the fused images of rgb format，And output (display includes that instinct shows and mobile display) on the display，To facilitate staff judge whether dangerous material according to the image that described display is shown and carry the people of dangerous material，Visible detection method i.e. can realize detection without stopping，Thus improve detection efficiency，Bigger Pedestrian flow detection can be met、Large range of detection and simultaneously multiple target being detected.
Accompanying drawing explanation
Fig. 1 is the structural representation of the hazardous material detection device of the present invention；
Fig. 2 is the flow chart of the hazardous material detection method of the present invention；
Fig. 3 is the flow chart of the hazardous material detection of the binaryzation gray level image of the present invention.
In figure, mark is illustrated as: 1-visible light image sensor；2-infrared image sensor；3-processor；4-video memory；5-display.
Detailed description of the invention
Below in conjunction with embodiment and accompanying drawing, technical scheme is further elaborated.
Embodiment 1
With reference to Fig. 1-3, present embodiments provide a kind of hazardous material detection device, including visible light image sensor 1, infrared image sensor 2, processor 3, video memory 4 and display 5；
Described visible light image sensor 1 signal is connected to described processor 3, to gather the visible images of the detected object comprising described dangerous material, and by the transmission of described visible images to described processor；
Described infrared image sensor 2 signal is connected to described processor 3, to gather the infrared image of the detected object comprising described dangerous material, and by described infrared image delivery to described processor；
Described visible images and infrared image are processed by described processor 3, the visible images after being processed and the infrared image after process；
Described video memory 4 signal is connected to described processor 3, with the infrared image after storing the visible images after described process and processing；The method that described visible images and infrared image process is included being compressed described visible images and infrared image by wherein said processor 3.
Described display 5 signal is connected to described processor 3, to show the image after described process.
nullThe hazardous material detection device of the present invention includes visible light image sensor 1、Infrared image sensor 2 and processor 3，The visible images of the raw RGB format that described visible light image sensor 1 is gathered is transferred to processor 3，The infrared image delivery of the raw RGB format that described infrared image sensor 2 is gathered is to processor 3，The visible images of described raw RGB format and the infrared image of raw RGB format are processed and merge by described processor 3，And the image after exporting described process by described display 5 and merge，With the infrared image of the definition and raw RGB format that make full use of the visible images of the described raw RGB format sensitiveness to the temperature of described dangerous material，Realize the detection to described dangerous material，This hazardous material detection device is after the infrared image of the visible images and raw RGB format that gather raw RGB format，Personnel's normal pass，Detection i.e. can be realized without stopping，Thus improve detection efficiency，Bigger Pedestrian flow detection can be met、Large range of detection and simultaneously multiple target being detected.
In the present embodiment, optionally, described display is the display of mobile terminal, such as mobile phone or display device for mounting on vehicle etc., and with when described hazardous material detection device is arranged at public place, staff can be in non-at-scene work.
In the present embodiment, optionally, described video memory 4 and/or display 5 are connected to described processor by WIFI wireless data transfer module, Bluetooth wireless data transport module or Zigbee wireless data transfer module, so that described video memory 4 and/or display 5 to be placed on different positions, thus facilitate staff that described display is monitored.
In the present embodiment, optionally, described video memory 4 includes local storage and/or cloud storage terminal, to realize described visible images and the storage of infrared image.
In the present embodiment, optionally, described processor 3 is FPGA, FMC or arm processor, and described hazardous material detection device also includes warning device, described warning device signal is connected to described processor 3, when described infrared image sensor 2 detects dangerous material, to report to the police.
In the present embodiment, optionally, described dangerous material include the object that gun, controlled knife etc. are different from human body temperature, to give full play to the described infrared image sensor sensitiveness to temperature, to divide human body and described dangerous material in described infrared image intermediate energy region.
Embodiment 2
With reference to Fig. 1-3, present embodiments provide a kind of method utilizing described hazardous material detection device detection dangerous material, comprise the following steps:
S10, utilize described visible light image sensor 1 and infrared image sensor 2 that the detected object comprising described dangerous material is carried out IMAQ, respectively obtain the visible images of raw RGB format and the infrared image of raw RGB format；To capture real-time coloured image by described visible light image sensor 1, i.e. obtain the visible images of raw RGB format；Described infrared image sensor 2 captures real-time coloured image, i.e. obtains the infrared image of raw RGB format；And 2-4 times of the resolution ratio that resolution ratio is described infrared image sensor 2 of described visible light image sensor 1, the resolution ratio of the visible images of the most described raw RGB format be described raw RGB format 2-4 times of infrared image resolution ratio, to facilitate follow-up image calibration to process with image co-registration；
S20, the visible images of raw RGB format is converted to the visible images of YIQ form；By from RGB to YIQ colour space transformation, the color information of the visible images of raw RGB format is separated with monochrome information, merge for successive image and prepare, and in the visible images of the YIQ form in the present invention, representative image brightness (luminance) is independent with the part of representative image color (chrominance), i.e. when the monochrome information of image individually being extracted and it being processed, can't the color information of the original image of this change；Vice versa.This feature of image just because of YIQ form, when detected dangerous material temperature information (information by entrained by the infrared image of described raw RGB format) is loaded in the visible images of raw RGB format, do not change other any information, such as color of original visible images；To improve the effect of described hazardous material detection method；
S30, generate visible ray gray level image according to the visible images of described YIQ form；Monochrome information (luminance) in the visible images of YIQ form is individually extracted by this step, generates one-dimensional visible ray gray level image；Monochrome information in the visible images of YIQ form will individually extract, and generate the gray level image of correspondence；To facilitate follow-up image co-registration to process；
S40, the infrared image of raw RGB format is converted into infrared hybrid optical system；To facilitate follow-up image co-registration to process；
S50, described infrared hybrid optical system is carried out pixel value reversion, the infrared hybrid optical system after being inverted；With by the gray level image after described reversion being merged the image after described fusion, regulation merge after the brightness of image, prevent the image after merging from crossing bright or the darkest, make the image after fusion consistent with the holding of visible ray gray level image as far as possible；
S60, the infrared hybrid optical system after described visible ray gray level image, infrared hybrid optical system and reversion is carried out image co-registration, the image after being merged；Image after described fusion is one-dimensional gray level image, and only includes new monochrome information；It is highly preferred that described image co-registration process uses linearity blending algorithm or Image Fusion based on wavelet transform etc..
S70, by after described fusion image generate YIQ form fused images；Monochrome information in the fused images of described YIQ form is new monochrome information, i.e. contains the temperature information of detected dangerous material, but does not change the color information of the visible images of original rgb format.
S80, the fused images of described YIQ form is converted to the fused images of rgb format；
S90, the fused images of described rgb format is shown on the display；
To be realized the detection to described dangerous material by above-mentioned steps, be can be seen that by above-mentioned steps, the hazardous material detection method of the present invention passes through RGB to YIQ colour space transformation, the monochrome information of described image is extracted, and monochrome information i.e. corresponds to the information such as position and the size of dangerous material, i.e. can accurately identify dangerous material by this operation and carry the people of dangerous material, visible detection method i.e. can realize detection without stopping, thus improve detection efficiency, bigger Pedestrian flow detection can be met, large range of detection and simultaneously multiple target being detected.
In the present embodiment, optionally, after described step S10, before step S20, also include step S15:
S15, visible images and original infrared image to raw RGB format carry out calibration process and frame synchronization process so that the pixel of the visible images of described raw RGB format and the infrared image of described raw RGB format and content matching；The purpose of described step S15 is the infrared image content matching of visible images and the raw RGB format making described raw RGB format, and synchronization is input to processor 3 and carries out subsequent treatment, to improve the described processor 3 treatment effect to the visible images of described raw RGB format with the infrared image of described raw RGB format, and the degree of accuracy that the position of dangerous material is identified.More preferably, the process that image calibration processes can use the image calibration algorithm of maturation, such as based on the image calibration algorithm that phase place is relevant, the image calibration algorithm of feature based conversion (scale invariant features transform), image calibration algorithm based on edge matching and multi-scale wavelet transformation etc..Frame synchronization process carries frame synchronization module by image processing hardware and realizes, it is possible to by hardware is programmed realization, such as based on VHDL FPGA hardware programs.
In the present embodiment, optionally, described step S80 and S90 are replaced with:
S100, the infrared image of raw RGB format is converted into the infrared image of YIQ form, prepares for extracting described dangerous material；
S110, the fused images of YIQ form and the infrared image of YIQ form are carried out image co-registration, obtain final fused images, it is highly preferred that described Image Fusion is linearity blending algorithm or Image Fusion based on wavelet transform etc.；Merge with the fused images by the infrared image of the YIQ form with temperature information with YIQ form；
S120, described final fused images is converted into final gray level image, to facilitate position and the size differentiating and extracting described dangerous material；
S130, described final gray level image is carried out binary conversion treatment, obtain binaryzation gray level image；
S140, by the visual image fusion of the white portion in described binaryzation gray level image Yu described raw RGB format, obtain image to be output；Will useful information in described binaryzation gray level image, during i.e. white portion is fused to the visible images of raw RGB format, and ensure that this white portion size position in the visible images of raw RGB format does not all change；
S150, product information the most dangerous in described binaryzation gray level image is judged；
S160, when, in step S150, not finding dangerous material, the visible images of raw RGB format is shown on described display 5；
S170, when in step S150, find dangerous material, described image to be output shown on described display 5.
To be chosen by intelligent target, the information of dangerous material is shown in visible images, colour space transformation to the infrared image of described raw RGB format, the information such as position and size of extracting described dangerous material, facilitate staff to decide whether described dangerous material carrier and process.
In the present embodiment, optionally, described step S130 specifically includes following steps:
S1301, setting threshold value, for the selection of dangerous material；
S1302, will be greater than being set to 1 equal to the value of the pixel of described threshold value；
S1303, the value of the pixel less than described threshold value is set to 0, obtains binaryzation gray level image；
So that described final gray level image is carried out binary conversion treatment, add the contrast of pixel corresponding to dangerous material position and size, facilitate staff to differentiate.
In the present embodiment, optionally, described step S150 specifically include following steps:
S1501, when the gray level image after described binaryzation does not exist the pixel that value is 1, then be judged to not find dangerous material；
S1502, when in the gray level image after described binaryzation existence value be the pixel of 1, then be judged to find dangerous material；To realize the most dangerous product in described binaryzation gray level image are carried out automatic decision.
In the present embodiment, optionally, described detection method also includes step S180；
S180, when in step S150, find dangerous material, report to the police.
In the present embodiment, optionally, the setting threshold value in described step S1301, could be arranged to 128；Or setting according to concrete application scenario manual debugging, the principle of setting selects suspicious object exactly, gets rid of and do not have value information simultaneously；Certain any system does not all have the intelligence in absolute sense, and in the case of there is leakage choosing and falsely dropping, the principle of threshold value setting is would rather to falsely drop to leak choosing；Or threshold value setting can also use the adaptive algorithm of image dualization, such as Back Propagation algorithm.
In the present embodiment, optionally, described step S140 specifically includes following steps:
S1401, the gray level image after binaryzation is carried out three-dimensional expansion；Image information that will be two-dimentional carries out duplication and becomes three-dimensional.
S1402, the pixel that value is 1 of described pixel is replaced the pixel of the visible images of the raw RGB format of same position, and the value of the pixel of the visible images of the raw RGB format of described same position is set to (1,1,1), to show the size and location of described dangerous material in described image to be output, staff is facilitated to differentiate.
The sequencing of above example only for ease of describing, does not represent the quality of embodiment.
Last it is noted that above example is only in order to illustrate technical scheme, it is not intended to limit；Although the present invention being described in detail with reference to previous embodiment, it will be understood by those within the art that: the technical scheme described in foregoing embodiments still can be modified by it, or wherein portion of techniques feature is carried out equivalent；And these amendments or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (4)
1. the method utilizing hazardous material detection device detection dangerous material, described hazardous material detection device includes
Visible light image sensor, infrared image sensor, processor, video memory and display；
Described visible light image sensor signal is connected to described processor, comprises described dangerous material to gather
The visible images of detected object, and by the transmission of described visible images to described processor；
Described infrared image sensor signal is connected to described processor, to gather the quilt comprising described dangerous material
The infrared image of detection object, and by described infrared image delivery to described processor；
Described visible images and infrared image are processed by described processor, the visible ray after being processed
Infrared image after image and process；
Described video memory signal is connected to described processor, to store the visible images after described process
With the infrared image after process；
Described monitor signal is connected to described processor, to show the image after described process；
Described video memory and/or display are by WIFI wireless data transfer module, Bluetooth wireless data
Transport module or Zigbee wireless data transfer module are connected to described processor；
It is characterized in that, comprise the following steps:
S10, utilize described visible light image sensor and the infrared image sensor quilt to comprising described dangerous material
Detection object carries out IMAQ, respectively obtains visible images and the raw RGB format of raw RGB format
Infrared image；
S15, the visible images of raw RGB format and the infrared image of raw RGB format are carried out at calibration
Reason and frame synchronization process so that the visible images of described raw RGB format and described raw RGB format
The pixel of infrared image and content matching；
S20, the visible images of raw RGB format is converted to the visible images of YIQ form；
S30, generate visible ray gray level image according to the visible images of described YIQ form；
S40, the infrared image of raw RGB format is converted into infrared hybrid optical system；
S50, described infrared hybrid optical system is carried out pixel value reversion, the infrared hybrid optical system after being inverted；
S60, the infrared hybrid optical system after described visible ray gray level image, infrared hybrid optical system and reversion is carried out
Image co-registration, the image after being merged；
S70, by after described fusion image generate YIQ form fused images；
S100, the infrared image of raw RGB format is converted into the infrared image of YIQ form；
S110, the fused images of YIQ form and the infrared image of YIQ form are carried out image co-registration, obtain
Final fused images；
S120, described final fused images is converted into final gray level image；
S130, described final gray level image is carried out binary conversion treatment, obtain binaryzation gray level image；
S140, by the visible ray of the white portion in described binaryzation gray level image Yu described raw RGB format
Image co-registration, and obtain image to be output；
S150, product information the most dangerous in described binaryzation gray level image is judged；
S160, when, in step S150, not finding dangerous material, by the visible images of raw RGB format in institute
State and show on display；
S170, when in step S150, find dangerous material, by described image to be output on the display
Display.
The method of detection dangerous material the most according to claim 1, it is characterised in that described step S130
Specifically include following steps:
S1301, setting threshold value；
S1302, will be greater than being set to 1 equal to the value of the pixel of described threshold value；
S1303, the value of the pixel less than described threshold value is set to 0 obtains binaryzation gray level image.
The method of detection dangerous material the most according to claim 2, it is characterised in that described step S140
Specifically include following steps:
S1401, the image after binaryzation is carried out three-dimensional expansion；Described three-dimensional expansion is by the image letter of two dimension
Breath carries out duplication and becomes three-dimensional；
S1402, the pixel that value is 1 of described pixel is replaced same position raw RGB format can
See the pixel of light image, and the pixel of the visible images by the raw RGB format of described same position
Value be set to (1,1,1).
The method of detection dangerous material the most according to claim 1, it is characterised in that described step S60
In the algorithm of image co-registration be: linearity blending algorithm or image co-registration based on wavelet transform are calculated
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410062083.7A CN103777252B (en) | 2014-02-24 | 2014-02-24 | Hazardous material detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410062083.7A CN103777252B (en) | 2014-02-24 | 2014-02-24 | Hazardous material detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103777252A CN103777252A (en) | 2014-05-07 |
| CN103777252B true CN103777252B (en) | 2016-08-24 |
Family
ID=50569711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410062083.7A Expired - Fee Related CN103777252B (en) | 2014-02-24 | 2014-02-24 | Hazardous material detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103777252B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106130633B (en) * | 2016-07-06 | 2019-03-12 | 深圳市检验检疫科学研究院 | Belongings infrared detection device and its image synchronization and storage method |
| WO2018072178A1 (en) * | 2016-10-20 | 2018-04-26 | 深圳达闼科技控股有限公司 | Iris recognition-based image preview method and device |
| CN109447071A (en) * | 2018-11-01 | 2019-03-08 | 博微太赫兹信息科技有限公司 | A kind of mm-wave imaging dangerous goods detection method based on FPGA and deep learning |
| CN109544494B (en) * | 2018-11-12 | 2021-08-17 | 北京航空航天大学 | Fusion method of passive millimeter wave image and visible light image in human body security check |
| CN112179497A (en) * | 2020-09-16 | 2021-01-05 | 安徽科杰粮保仓储设备有限公司 | AI technology-based campus personnel behavior abnormity early warning robot |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4532765B2 (en) * | 2001-03-13 | 2010-08-25 | キヤノン株式会社 | Image processing apparatus and image processing program |
| CN1932882A (en) * | 2006-10-19 | 2007-03-21 | 上海交通大学 | Infared and visible light sequential image feature level fusing method based on target detection |
| JP4608568B2 (en) * | 2008-07-04 | 2011-01-12 | キング通信工業株式会社 | Security monitoring device |
| CN102982518A (en) * | 2012-11-06 | 2013-03-20 | 扬州万方电子技术有限责任公司 | Fusion method of infrared image and visible light dynamic image and fusion device of infrared image and visible light dynamic image |
-
2014
- 2014-02-24 CN CN201410062083.7A patent/CN103777252B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN103777252A (en) | 2014-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103926199B (en) | Hazardous material detection method | |
| CN103900973B (en) | Hazardous material detection method | |
| CN103777252B (en) | Hazardous material detection method | |
| CA2640884C (en) | Methods and systems for use in security screening, with parallel processing capability | |
| Kong et al. | Detecting abandoned objects with a moving camera | |
| CN103646250B (en) | Pedestrian monitoring method and device based on distance image head and shoulder features | |
| CN110751079A (en) | Article detection method, apparatus, system and computer readable storage medium | |
| binti Zaidi et al. | Fire recognition using RGB and YCbCr color space | |
| US8922674B2 (en) | Method and system for facilitating color balance synchronization between a plurality of video cameras and for obtaining object tracking between two or more video cameras | |
| WO2017027212A1 (en) | Machine vision feature-tracking system | |
| CN107561737B (en) | Method, device and tool for automatically adjusting IPC (International Process control) parameters of AOI (automatic Option identification) system | |
| CN105869115B (en) | A kind of depth image super-resolution method based on kinect2.0 | |
| Yadav et al. | Optimized flame detection using image processing based techniques | |
| CN109409395A (en) | Using the method for template matching method identification target object region electrical symbol in power monitoring | |
| JP2017504017A (en) | Measuring instrument, system, and program | |
| EP3772722A1 (en) | X-ray image processing system and method, and program therefor | |
| JP2018169831A (en) | Apparatus for image comparison | |
| CN105554456B (en) | Method for processing video frequency and equipment | |
| Shweta | Intelligent refrigerator using artificial intelligence | |
| KR101509593B1 (en) | Image classification method and apparatus for preset tour camera | |
| Wang et al. | An improved YOLOX approach for low-light and small object detection: PPE on tunnel construction sites | |
| Poobalan et al. | Fire detection based on color filters and Bag-of-Features classification | |
| JP2017527905A (en) | Foreign matter detection system based on color | |
| CN105427303B (en) | A kind of vision measurement and method of estimation of substation's legacy | |
| CN112037255A (en) | Target tracking method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190329 Address after: 518109 C230, 1st floor, Hairong Hao Yuan, Dalangnan Road, Hualian Community, Longhua Street, Longhua District, Shenzhen City, Guangdong Province Patentee after: Shenzhen Hongwei Zhichuang Communication Technology Co., Ltd. Address before: 030024 Room 2603, Xigao, 53 Xikuangjie, Wanbailin District, Taiyuan City, Shanxi Province Patentee before: Huang Xiaopeng |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160824 Termination date: 20200224 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |