CN102707322B - Human body detection device based on millimeter wave imaging - Google Patents
Human body detection device based on millimeter wave imaging Download PDFInfo
- Publication number
- CN102707322B CN102707322B CN201210050292.0A CN201210050292A CN102707322B CN 102707322 B CN102707322 B CN 102707322B CN 201210050292 A CN201210050292 A CN 201210050292A CN 102707322 B CN102707322 B CN 102707322B
- Authority
- CN
- China
- Prior art keywords
- human body
- image
- module
- original image
- human
- 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.)
- Active
Links
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a human body detection device based on millimeter wave imaging. The human body detection device comprises a scanner, an adjusting module, a segmenting and positioning module and a human body model generation module. The scanner is used for millimeter wave scanning of personnel to be detected to acquire original images, the adjusting module is used for adjusting the original images to acquire target images, the segmenting and positioning module is used for segmenting and positioning human body parts according to the target images, and the human body model generation module is used for generating human body models. By the human body detection device based on millimeter wave imaging, human body recognition and processing in millimeter wave images are realized, and foundations are provided for concealed article inspection and privacy protection technique.
Description
Technical field
The present invention relates to the detection technique in safety check field, more specifically, the present invention relates to a kind of human body detection device based on mm-wave imaging.
Background technology
In safety check field, for the detection of human body and concealment article thereof, there is following various ways: metal detector, x-ray fluoroscopy, infrared detection and millimeter wave detection etc.Metal detector is to realize by electromagnetic induction, can only judge having or not of metal object, can not imaging or definite object space.X-ray fluoroscopy equipment has very strong penetrability, is generally used for the detection for luggage and articles, as directly can be larger to people's danger to human detection, therefore in safety check, be generally rarely used in human detection.Infrared detection is to utilize the thermal radiation property of object to carry out imaging, can be used for the detection to human body in safety check.The brightness of the object in infrared image depends primarily on the surface emissivity characteristic of the temperature of object and the heat of radiation and object, be characterized in not having significant corner angle, marginal information, its edge lines are round and smooth, and grey scale change is slow, and the shape details to object and small attitude change insensitive.These features make the human body in infrared image detect and have certain difficulty.
Millimeter wave (3GHz-300GHz) is a kind of electromagnetic wave between light wave and radiowave.Millimeter wave can penetrate all clothing clothes, and the millimeter wave energy of human body radiation is eager to excel in whatever one does compared with metal, pottery, plastic explosive, powder type explosive and clothing, insulating material etc., utilize active/passive millimeter-wave technology to detect to be hidden in the prohibited items such as the various cutters, gun, explosive of human body surface.Due to stronger than metal detection technical capability, more safer than ray technology, over nearly 10 years, human body millimeter wave safety check technology is rapidly developed.Focal plane arrays (FPA) scanning technique, multi-beam frequency-scan technique and the active 3D hologram millimeter-wave technology of passive-type obtain test and application in succession.After utilizing active MMW rays safety detection apparatus to human body imaging, in image, can show more clearly the various article that characteristics of human body and human body carry.
First,, in millimeter wave safety check, the analysis of human body image is important composition link.After human body mm-wave imaging, how human body image being detected to analysis, is the basis that safe examination system realize target detects robotization, is to the concealment sign of article position on human body and the basis to human body image secret protection in subsequent treatment.
Secondly, after mm-wave imaging, how concealment article are detected and sign on human body, it in prior art, is the method by manual analysis, wherein image enhancement technique and multiframe comparison techniques are applied in manual analysis, but need to be by professional person's interpretive analysis, can realize identification and location to concealment article.Although the image Segmentation Technology based on methods such as the many threshold values of gray scale, Boundary Extraction, rim detection, Region Segmentation, wavelet transformation, morphology, fuzzy mathematics, genetic algorithm, neural network, information entropys is attempted and is applied in the automatic detection of concealment article, but owing to disconnecting mutually with human vision mechanism, only utilize gray scale in image and spatial information to Image Segmentation Using, still can produce and the inconsistent situation of human vision.And method for positioning analyzing based on human body prior model, in the motion tracking of human body, be applied, reduce the complexity of following the tracks of, wherein mainly comprise bar band model, the bar-shaped model of Figure 33 etc. shown in figure 32, but because stick model only comprises human body contour outline information, as structure, shape, attitude etc., and bar-shaped model only comprises each articulation point of human body, all can only be limited to the detection of human body, still can not directly solve the automatic detection and Identification problem of concealment article at human body.
The 3rd, pass through millimeter wave scanning imaging, concealment Item Information on can human body, but can cause exposure and the demonstration of human body privacy (as face and privacy places) simultaneously, after mm-wave imaging, how image is carried out to analyzing and processing, show that the privacy information of the front shielding of concealment article human body is also a technical matters that needs solution in safe examination system.
Summary of the invention
The object of the invention is to provide a kind of human body detecting method and device based on mm-wave imaging, realizes identification and location to partes corporis humani position in millimeter wave scanning.
The method comprises the following steps: tested personnel are carried out to millimeter wave scanning and obtain original image; Described original image is adjusted and obtained target image; Carry out cutting apart of human body and locate according to described target image; Generate manikin.
Further, describedly carry out cutting apart of human body and locate also comprising following sub-step according to described target image: the vertical center line of determining human body; Determine the coordinate of the each key point of described target image human body and obtain the horizontal cut-off rule between partes corporis humani position; Determine width and the slope of partes corporis humani position.
Further, described generation manikin comprises: according to width and the slope of the coordinate of described each key point, described partes corporis humani position, obtain the manikin with rectangle and/or parallelogram composition.
Further, described to described original image adjust obtain target image also comprise following sub-step: described original image is carried out to pre-service and obtains preliminary denoising image; Described preliminary denoising image is carried out to binaryzation and obtain preliminary bianry image; Described preliminary bianry image is processed and obtained described target image.
Further, describedly described original image is carried out to pre-service obtain preliminary denoising image and further comprise following sub-step: described original image and background image gray-scale value carry out difference computing; Picture smooth treatment; Linear greyscale transformation.
Further, describedly described preliminary denoising image is carried out to binaryzation to obtain preliminary bianry image be to utilize Pulse Coupled Neural Network algorithm to be criterion to the maximum and to choose the threshold value of binaryzation with entropy.
Further, described to described preliminary bianry image process again obtain described target image be to pass through morphologic filtering.
Further, describedly comprise by morphologic filtering: use square structure element that the length of side is 5 to carry out erosion operation and eliminate the bright noise spot outside human body; The square structure element that the use length of side is 4 carries out eliminating when opening operation keeps image size isolated area and the burr at human body edge; When carrying out closed operation maintenance image size, the square structure element that the use length of side is 4 fills the tiny cavity in human body and the edge of level and smooth human body; The square structure element that the use length of side is 5 carries out dilation operation makes image return to life size.
Further, describedly described original image is carried out to pre-service obtain preliminary denoising image and also comprise following sub-step: described original image is carried out to figure image intensifying.
Correspondingly, the human body detection device based on mm-wave imaging of the present invention, comprising: scanister, obtains original image for tested personnel being carried out to millimeter wave scanning; Adjusting module, obtains target image for described original image is adjusted; Cut apart locating module, for carrying out cutting apart of human body and locate according to described target image; Manikin generation module, for generating manikin.
Further, described in cut apart locating module and also comprise following submodule: vertically center line module, for determining the vertical center line of human body; Coordinate horizontal line module, for determining the coordinate of the each key point of described target image human body and obtaining the horizontal cut-off rule between partes corporis humani position; Width slope module, for determining width and the slope of partes corporis humani position.
Further, described manikin generation module is further used for according to width and the slope of the coordinate of described each key point, described partes corporis humani position, obtains the manikin with rectangle and/or parallelogram composition.
Further, described adjusting module also comprises following submodule: pretreatment module, obtains preliminary denoising image for described original image being carried out to pre-service; Binarization block, obtains preliminary bianry image for described preliminary denoising image is carried out to binaryzation; Processing module again, obtains described target image for described preliminary bianry image is processed again.
Further, described pretreatment module further comprises with lower unit:
Difference arithmetic element, for carrying out difference computing by described original image and background image gray-scale value; Smoothing processing unit, for carrying out picture smooth treatment; Linear change unit, for carrying out linear greyscale transformation.
Further, described binarization block further utilizes Pulse Coupled Neural Network algorithm to be criterion to the maximum and to choose the threshold value of binaryzation with entropy.
Further, described processing module is more further processed by morphologic filtering again.
Further, describedly comprise by morphologic filtering: use square structure element that the length of side is 5 to carry out erosion operation and eliminate the bright noise spot outside human body; The square structure element that the use length of side is 4 carries out eliminating when opening operation keeps image size isolated area and the burr at human body edge; When carrying out closed operation maintenance image size, the square structure element that the use length of side is 4 fills the tiny cavity in human body and the edge of level and smooth human body; The square structure element that the use length of side is 5 carries out dilation operation makes image return to life size.
Further, described pretreatment module also comprises: image enhancing unit, and for described original image is carried out to figure image intensifying.
By human body detecting method and the device based on mm-wave imaging of the present invention, realize the identification to human body parts and processing in millimeter-wave image, for subsequent survey concealment article and secret protection provide the foundation.
Location and identification that the object of the invention is also to provide a kind of automatic detection and Identification method and apparatus of hiding article to realize and in millimeter wave scanning, concealment article is distributed on human body are automatic from manually becoming, and reduce personnel's request for utilization.
The automatic detection and Identification method of described concealment article, comprises the following steps: tested personnel are carried out to millimeter wave scanning and obtain original image; Described original image is adjusted and obtained target image; Carry out cutting apart of human body and locate according to described target image; Generate bar rod combination model; According to described original image, non-human target is detected, obtain non-human target distribution original image; Utilize described excellent combination model to obtain the location distribution information of described non-human target distribution original image with respect to human body; Described non-human target is carried out classification identification and shown the location distribution information of concealment article with respect to human body.
Further, described generation bar rod combination model comprises following sub-step: generate the bar-shaped model that the each key point of human body is provided; Generation provides the bar band model of human body contour outline information; In conjunction with described bar-shaped model and described band model, generate bar rod combination model.
Further, describedly according to described original image, non-human target is detected, obtain non-human target distribution original image and comprise following sub-step: described original image is carried out to rim detection, tentatively identify non-human target; Highlight non-human target distribution region by mathematical morphological operation; Choose minimum circumscribed rectangle according to the border in described non-human target distribution region and obtain non-human goal rule regional distribution chart; Merge described non-human goal rule regional distribution chart and described original image, obtain described non-human target original image.
It is further, described that to utilize described excellent combination model to obtain described non-human target distribution original image be by described non-human target distribution original image is input on described excellent combination model with respect to the location distribution information of human body.
Further, described to described non-human target carry out classification identification and show concealment article comprise following sub-step with respect to the location distribution information of human body: the exposed position of human body is positioned; The non-human target being distributed on the exposed position of human body is defined as to non-concealment article, the non-human target being distributed in outside the exposed position of human body is defined as hiding article; Reject the original image of described non-concealment article and show the distributed intelligence of concealment article original image on described excellent combination model.
Correspondingly, the automatic detection and Identification device of concealment article of the present invention, comprising: scanister, obtains original image for tested personnel being carried out to millimeter wave scanning; Adjusting module, obtains target image for described original image is adjusted; Cut apart locating module, for carrying out cutting apart of human body and locate according to described target image; Bar rod combination model generation module, for generating bar rod combination model; Non-human target Preliminary detection module, for non-human target being detected according to described original image, obtains non-human target distribution original image; Non-human target distribution module, for utilizing described excellent combination model to obtain the location distribution information of described non-human target distribution original image with respect to human body; Classification recognition module, for carrying out classification identification and showing the location distribution information of concealment article with respect to human body to described non-human target.
Further, described excellent combination model generation module comprises following submodule: bar-shaped model generation module, for generating the bar-shaped model that the each key point of human body is provided; Band model generation module, provides the bar of human body contour outline information band model for generating; Binding modules, in conjunction with described bar-shaped model and described band model.
Further, described non-human target Preliminary detection module comprises following submodule: rim detection module, for described original image is carried out to rim detection, tentatively identify non-human target; Highlight module, for highlight non-human target distribution region by mathematical morphological operation; Regularization module, obtains non-human goal rule regional distribution chart for choosing minimum circumscribed rectangle according to the border in described non-human target distribution region; Fusion Module, for merging described non-human goal rule regional distribution chart and described original image, obtains described non-human target original image.
Further, described non-human target distribution module is by described non-human target distribution original image is input on described excellent combination model.
Further, described classification recognition module comprises following submodule: exposed spots localization module, for the exposed position of human body is positioned; Sort module, for the non-human target being distributed on the exposed position of human body is defined as to non-concealment article, is defined as hiding article by the non-human target being distributed in outside the exposed position of human body; Display module, for rejecting the original image of described non-concealment article and showing the distributed intelligence of concealment article original image on described excellent combination model.
The method and apparatus of the automatic detection and Identification by concealment article of the present invention, has realized the detection and Identification of concealment article from manually becoming automatically, has reduced personnel's request for utilization, has reduced personal error, has shortened the detection interpretation time.
The object of the invention is also to provide a kind of method for secret protection and device based on mm-wave imaging, has realized the secret protection to tested personnel in millimeter wave scanning.
Method for secret protection based on mm-wave imaging of the present invention, comprises the following steps: tested personnel are carried out to millimeter wave scanning and obtain original image; Carry out human detection and concealment Articles detecting according to described original image; Determine the privacy places of human body; The privacy places of human body is shielded and indicates the concealment Item Information on human body.
Further, the privacy places of described definite human body comprises: the sex that judges tested personnel, in the time that tested personnel are the male sex, the region of head zone and the downward trunk width 1/2 of human body waist is defined as privacy places, in the time that tested personnel are women, the region of human body head region, the downward trunk width 1/2 of human body waist and trunk are privacy places from trunk upper end down to the region of trunk height 1/2.
Further, the concealment Item Information that the described privacy places to human body shields and indicates on human body comprises: on described original image, privacy places is carried out to Fuzzy processing forming section obfuscation original image; On described part obfuscation original image, described concealment article are marked with sign frame.
Further, the concealment Item Information that the described privacy places to human body shields and indicates on human body comprises: select the object image in described human detection; Judge the whether privacy places in human body of concealment article, if so, use the color blocks different from human body color represent to hide article and indicate on described object image; If not, the original image of concealment article is presented on described object image.
Further, the concealment Item Information that the described privacy places to human body shields and indicates on human body comprises: on described original image, human body is carried out to whole Fuzzy processing and form whole obfuscation original images; Judge the whether privacy places in human body of concealment article, if so, use the color blocks different from human body color represent to hide article and indicate on described whole obfuscation original images; If not, the original image of concealment article is presented on described whole obfuscation original image.
Further, the concealment Item Information that the described privacy places to human body shields and indicates on human body comprises: select the manikin in described human detection; Use the color blocks different from manikin color represent to hide article and indicate on described manikin.
Correspondingly, the secret protection device based on mm-wave imaging of the present invention comprises: scanister, obtains original image for tested personnel being carried out to millimeter wave scanning; Pick-up unit, for carrying out human detection and concealment Articles detecting according to described original image; Privacy places determination module, for determining the privacy places of human body; Privacy mask module, for shielding and indicate the concealment Item Information on human body to the privacy places of human body.
Further, described privacy places determination module is further used for: the sex that judges tested personnel, in the time that tested personnel are the male sex, the region of head zone and the downward trunk width 1/2 of human body waist is defined as privacy places, in the time that tested personnel are women, the region of human body head region, the downward trunk width 1/2 of human body waist and trunk are privacy places from trunk upper end down to the region of trunk height 1/2.
Further, described privacy mask module further comprises following submodule: part obfuscation module, for privacy places being carried out to Fuzzy processing forming section obfuscation original image on described original image; First indicates module, for described concealment article being marked with sign frame on described part obfuscation original image.
Further, described privacy mask module further comprises: select object image module, for selecting the object image of described human detection; Second indicates module, for judging the whether privacy places in human body of concealment article, if so, uses the color blocks different from human body color represent to hide article and indicate on described object image; If not, the original image of concealment article is presented on described object image.
Further, described privacy mask module further comprises: all obfuscation modules, form whole obfuscation original images for human body being carried out to whole Fuzzy processing on described original image; The 3rd indicates module, for judging the whether privacy places in human body of concealment article, if so, uses the color blocks different from human body color represent to hide article and indicate on described whole obfuscation original images; If not, the original image of concealment article is presented on described whole obfuscation original image.
Further, described privacy mask module further comprises: preference pattern module, for selecting the manikin of described human detection; The 4th indicates module, for using the color blocks different from manikin color to represent to hide article and indicating on described manikin.
By method for secret protection and the device based on mm-wave imaging of the present invention, the exposure to human body privacy while having avoided detecting concealment article, has realized the effective protection to privacy places of human body.
Brief description of the drawings
Below with reference to accompanying drawings and in conjunction with the embodiments the present invention is specifically described.
Fig. 1 is the human body detecting method basic flow sheet based on mm-wave imaging;
Fig. 2 is the human body detection device basic block diagram based on mm-wave imaging;
Fig. 3 is original image;
Fig. 4 is preliminary denoising image;
Fig. 5 is preliminary bianry image;
Fig. 6 is the process flow diagram of step S2 in the human body detecting method based on mm-wave imaging;
Fig. 7 is the structural representation of adjusting module in the human body detection device based on mm-wave imaging;
Fig. 8 is object image;
Fig. 9 is the key figure of human body;
Figure 10 is human body segmentation effect figure;
Figure 11 is the process flow diagram of step S3 in the human body detecting method based on mm-wave imaging;
Figure 12 is the structural representation of cutting apart locating module in the human body detection device based on mm-wave imaging;
Figure 13 is the manikin figure obtaining in human body detecting method based on mm-wave imaging and device;
Figure 14 is manikin figure and the corresponding design sketch of original image;
Figure 15 is the basic flow sheet of the automatic detection and Identification method of concealment article;
Figure 16 is the process flow diagram of step S5 in the automatic detection and Identification method of concealment article;
Figure 17 is bar rod combination model figure;
Figure 18 is the process flow diagram of step S6 in the automatic detection and Identification method of concealment article;
Figure 19 is the non-human target image of preliminary identification;
Figure 20 is non-human target distribution areal map;
Figure 21 is non-human goal rule regional distribution chart;
Figure 22 is non-human target original image;
Figure 23 is the distribution plan of non-human target original image on bar rod combination model;
Figure 24 is the distribution plan of concealment article original image on bar rod combination model;
Figure 25 is the process flow diagram of step S8 in the automatic detection and Identification method of concealment article;
Figure 26 is the basic flow sheet of the method for secret protection based on mm-wave imaging;
Figure 27 is the structural representation of the secret protection device based on mm-wave imaging;
Figure 28 is the design sketch of the first embodiment of privacy mask;
Figure 29 is the design sketch of the second embodiment of privacy mask;
Figure 30 is the design sketch of the third embodiment of privacy mask;
Figure 31 is the design sketch of the 4th kind of embodiment of privacy mask;
Figure 32 is the bar band model schematic diagram in background technology;
Figure 33 is the bar-shaped model schematic diagram in background technology.
Embodiment
With reference to the accompanying drawings and by embodiments of the invention, technical scheme of the present invention is described in detail.
A kind of human body detecting method based on mm-wave imaging in the present invention, comprises the following steps: S1, tested personnel are carried out to millimeter wave scanning obtain original image; S2, to described original image adjust obtain target image; S3, carry out cutting apart of human body and locate according to described target image; S4, generation manikin.As shown in Figure 1.
Correspondingly, as shown in Figure 2, the present invention also provides a kind of human body detection device based on mm-wave imaging, comprising:
Scanister 1, for performing step S1, carries out millimeter wave scanning to tested personnel and obtains original image;
Adjusting module 2, for performing step S2, adjusts and obtains target image described original image;
Cut apart locating module 3, for performing step S3, carry out cutting apart of human body and locate according to described target image;
Manikin generation module 4, for performing step S4, generates manikin.
In step S1, require tested personnel to enter millimeter wave scan test section, by the mode of scanister 1 millimeter wave active/passive scan detect after obtain original image as shown in Figure 3.Original image after scanning generally has following features: integral image is clear not, comprises much noise.
Therefore need adjusting module 2 to carry out step S2, thereby original image is adjusted and obtained the target image that is suitable for carrying out image operation and cuts apart, as Fig. 6, step S2 comprises following sub-step: S21, original image is carried out to pre-service obtains preliminary denoising image; S22, preliminary denoising image is carried out to binaryzation obtain preliminary bianry image; S23, to preliminary bianry image process again obtain described target image.
Correspondingly, as Fig. 7, adjusting module 2 also comprises following submodule:
Pretreatment module 21, for performing step S21, carries out pre-service to described original image and obtains preliminary denoising image;
Binarization block 22, for performing step S22, carries out binaryzation to described preliminary denoising image and obtains preliminary bianry image;
Processing module 23 again, for performing step S23, processes and obtains described target image described preliminary bianry image.
Further, pretreatment module 21 also comprises image enhancing unit, difference arithmetic element, smoothing processing unit, linear change unit.Pretreatment module 21 performs step S21 need to carry out following sub-step:
Image enhancing unit makes original image human region and background area Enhancement contrast for original image is carried out to figure image intensifying, improves image visual effect.
Difference arithmetic element for undertaken by the gray-scale value of original image and empty background image difference computing in other words the gray-scale value of original image and empty background image subtract each other, thereby elimination system noise.Empty background image is exactly while thering is no tested personnel in millimeter wave scan test section, to scan formed image.
Smoothing processing unit removes the random noise of image for carrying out the smoothing processing of image, pass through to use in this unit
operator carries out low-pass filtering to image and realizes smooth operation.
Linear change unit is for carrying out gray scale stretching or claiming subregion linear transformation to image, tonal range to uninterested background area in image is compressed, human region tonal range is expanded, thereby outstanding human body parts, make human body parts entirety clear, finally obtain preliminary denoising image, as shown in Figure 4.
Further, the performed step S22 of binarization block 22, preliminary denoising image is carried out to binaryzation, and to obtain preliminary bianry image be to utilize Pulse Coupled Neural Network (PCNN) algorithm to be criterion to the maximum and to choose the threshold value of binaryzation with entropy, utilize this threshold value the gray-scale map of preliminary denoising image to be converted to the image of binaryzation, thereby realize cutting apart of human region and background area in image.
How by selected threshold, human body and background more completely being cut apart is accurately the key of problem, here use comparatively ripe Pulse Coupled Neural Network (PCNN) technology, PCNN be that the nineties, Eckhorn etc. researched and proposed based on mammiferous visual characteristics such as cats based on Pulse-coupled Neural Network Model, the iterative process that this model is chosen for the threshold value of image, iterative formula is as follows:
F is exactly i, j neuronic n feed back input F
i, j[n], I
i, j[n] is that input stimulus signal (gray-scale value of i, a j pixel in the matrix forming for image pixel here), β are coefficient of connection, L
i, j[n] connects item, T
i, j[n] is dynamic threshold, the threshold value that will solve in the present invention, Y
i, j[n] is PCNN pulse output valve, U
i, j[n] is internal activity item.The wherein M of inner connection matrix M, W (general W=M)
i, j, k, l, W
i, j, k, lbe respectively F
i, j[n], L
i, jy in [n]
i, jthe weighting coefficient of [n], α
f, α
l, α
tbe respectively F
i, j[n], L
i, j[n], T
i, jthe damping time constant of [n], V
f, V
l, V
tbe respectively F
i, j[n], L
i, j[n], T
i, jintrinsic electromotive force in [n].
Entropy is a kind of form of expression of image statistics characteristic, has reflected the size of image inclusion information amount.For image, generally cut apart rear Image entropy larger, illustrate that to obtain quantity of information from former figure after cutting apart larger, cut apart image detail abundanter, thereby overall segmentation effect also should be better.This patent uses entropy to be the criterion that criterion finishes as PCNN iteration to the maximum.The computing formula of entropy is:
H
1(P)＝-P
1×log
2P
1-P
0×log
2P
0。
Wherein, P
1, P
0respectively indicating impulse output valve Y[n] be 1, be 0 probability.The present invention, by setting a very large iterations n, as n=100, uses PCNN algorithm to carry out interative computation, after each computing finishes, obtains corresponding entropy H
1(P), the entropy that then relatively n computing obtains, obtains the wherein value H of entropy maximum
1max(P) iterations N time
max.Iterations is N
maxtime obtained threshold value T[N
max], the Y[N of now PCNN output
max] form under other parameters one stable condition of PCNN the bianry image of overall segmentation effect the best.Wherein Y[N
max] be that 1 part is background parts, Y[N
max] be that 0 part is human body parts.
The span that is adapted to each parameter in the PCNN formula of above-mentioned computation process is:
| α F | α L | α T | V F | V L | V T | β |
| 0.1～0.6 | 1～10 | 0.1～0.6 | 0.1～0.5 | 0.1～0.5 | 2～10 | 0.1～0.6 |
W, two operators of M can use 1/r or 1/r
2element form form matrix, r represents the matrix length of side of operator.
Preferably, the present invention can get following parameter value: α
f=0.2, α
l=2, α
t=0.1, V
f=0.1, V
l=0.5, V
t=20, β=0.5,
Calculate, the preliminary bianry image of the optimum efficiency of acquisition as shown in Figure 5.
Further, the more performed step S23 of processing module 23, described preliminary bianry image being processed and obtained described target image, is to pass through morphologic filtering.Because Threshold segmentation can cause picture noise, this noise is mainly the isolated dark noise point in isolated bright noise spot or the human body outside human body parts.In order to remove these noises, the method that the present invention uses is that preliminary bianry image applied mathematics morphology operations method is carried out to filtering and conversion, obtaining an image with the binaryzation of clear level and smooth profile is described target image, as shown in Figure 8, thereby is beneficial to follow-up processing.
Mathematical morphological operation method mainly comprises erosion operation, dilation operation, opening operation and closed operation.
Erosion operation can weaken even to be eliminated and is less than structural element bright areas, thereby can be used for effectively removing rough projection on isolated noise point border.
Dilation operation is that all background dots that contact with target object are incorporated into the process in object, can filling cavity and form together with rough sunk part in territory and flat image boundary.
Opening operation is first image to be carried out to erosion operation to carry out dilation operation again, can get rid of isolated area and burr in image, utilization can be eliminated the noise spot that shape is less than structural element, according to the feature of target noise, select suitable structural element, just can reject target noise, and background is remained.
Closed operation is first image to be carried out to dilation operation to carry out erosion operation again, can fill the tiny cavity in object, connects contiguous thing and level and smooth object boundary.
Wherein, structural element is the basic operator of mathematical morphological operation, and the structural element of choice for use is mainly the shape and size of structural element.
Preferably, the morphology operations that this method is used can be following processing procedure: (1) is used the square structure element that the length of side is 5 to carry out the bright noise spot outside human body in erosion operation removal of images to image; (2) use the square structure element that the length of side is 4 to carry out opening operation to image, keep image size to eliminate isolated area and the burr at human body edge simultaneously; (3) use the square structure element that the length of side is 4 to carry out closed operation to image, when keeping image size, fill the tiny cavity in human body, and the border of level and smooth human body; (4) the square structure element that the use length of side is 5 carries out expansive working to image makes image return to life size.Can remove long and wide 5 the noise that is all less than by this process, fill up long on human body and wide 5 the cavity that is all less than, in the target image forming after finishing dealing with, comprise an approximate complete human body parts, make characteristics of human body more obvious.
In addition, if the region that area is less can, by the area of each connected region in computed image, be removed in the White lnterfere region that also has area not to be eliminated greatly in image.
Further, as Figure 11, step S3, carry out cutting apart of human body and locate the vertical center line that also comprises following sub-step: S31, determines human body according to described target image; S32, determine the coordinate of the each key point of described target image human body and obtain the horizontal cut-off rule between partes corporis humani position; S33, the width of determining partes corporis humani position and slope.
Correspondingly, as shown in figure 12, cut apart locating module 3 and also comprise following submodule:
Vertically center line module 31, for performing step 31, determine the vertical center line of human body;
Coordinate horizontal line module 32, for performing step 32, determine the coordinate of the each key point of described target image human body and obtaining the horizontal cut-off rule between partes corporis humani position;
Width slope module 33, for performing step 33, determine width and the slope of partes corporis humani position.
In the time that vertically center line module 31 is determined people's the position of vertical center line in performing step S31, because the human region in target image has symmetrical property, therefore calculate the total pixel of the image of this human region and, as use S
0represent, then from human region left side edge start by the row of image calculate from left to right human body parts image pixel and, as use S
1represent, work as S
1for S
01/2 o'clock, when prostatitis is the vertical center line of human body.
Perform step in S32 in coordinate horizontal line module 32, the coordinate of the each key point of human body is exactly partes corporis humani's bit position coordinate, as coordinate of marginal end point coordinate, central point etc., human body comprises: the crown, sole, neck, trunk upper end, trunk lower end (waist), crotch, knee, finger tip and ancon.Horizontal cut-off rule between the coordinate of the each key point of human body and each position is the process of mutually calculating, specific as follows:
Vertical center line along human body is downward from image top, the first man body region frontier point finding, if judge from this point and continue to have downwards continuously and length is not less than 1/10 same gray-scale value of picture altitude, determine the crown central point that this point is human body, the crown horizontal line H2 that the residing horizontal line of this point is human body.
And the position of human foot is fixing position in image, therefore can determine that sole coordinate and residing horizontal line are H9.Because the position that when imaging, people stands is fixed, can obtain people's height H with the ordinate that the ordinate on the crown deducts sole.
According to human anatomy judgement, head part accounts for 15% of height, can determine coordinate and the residing horizontal line H4 of neck upper end.The height of neck accounts for 45% of height of head, therefore can determine coordinate and the residing horizontal line H5 of the edge end points of trunk upper end.
The represented target image of Fig. 8 is carried out to refinement and can also obtain the key figure of human body, as shown in Figure 9.
In the key figure of Fig. 9 human body, the point of crossing of trunk and two legs is as the position of trunk lower end, and its place horizontal line, as stringcourse H6, is the long H of leg along this line to the horizontal distance definition of pin
leg.
According to human anatomy judgement, the ratio of human body shank and thigh is about 1: 1.2, therefore can determine that the horizontal level of knee is from upwards H of pin
leg× 5/11 position, therefore obtains knee horizontal line H8.
On the target image shown in Fig. 8, from image bottom along vertical center line upwards, with first position of intersecting point of human body image as crotch, thereby obtain human body crotch horizontal line H7.
Taking vertical center line as cut-off rule, the image of human region is divided into left and right two halves, the peak of left one side of something is the position of left hand finger tip, and on image right, the peak of human body is exactly the position of right hand finger tip, thereby obtains finger tip horizontal line H0.In the present invention, the difference in height between two finger tips is ignored.
Due to before tested personnel are scanned, require its two arm outwards to open, therefore on target image, the width between two elbows is the widest position of human body, and the position of therefore finding the human region leftmost side is exactly the left elbow of health, the rightmost side is exactly the right elbow of health, thereby obtains elbow lever line H3.In the present invention, the difference in height between two elbows is ignored.
According to human dissection theory, the Length Ratio of general human body hand and upper arm is 7: 9, thus can determine the position of wrist according to hand point and the position of elbow, thus also obtain wrist horizontal line H1.
Human body segmentation effect figure as shown in figure 10.
In the performed step S33 of width slope module 33, according to the width of determining partes corporis humani position crossing with each horizontal cut-off rule of partes corporis humani position in the target image of Fig. 8, key point (as two central points at the two ends up and down of a part) coordinate of recycling partes corporis humani position calculates the slope at this position.
Further, the performed step S4 of manikin generation module 4 generates manikin and comprises respectively according to the coordinate of each position key point, width and slope, represent each position with rectangle or parallelogram, all positions are linked together, obtain a manikin with rectangle and/or parallelogram composition, as shown in figure 13.This human body can carry out proportional correspondingly with the human region in original image, and effect as shown in figure 14.
It should be noted that the manikin generating in the description of this method is the explanation of carrying out as an example of bar band model example.According to target image carry out cutting apart of human body and locate after can also generate bar-shaped model.And bar band model and bar-shaped model all can be applied in a kind of following automatic detection and Identification method of hiding article.
On the other hand, the present invention also provides a kind of automatic detection and Identification method and a kind of automatic detection and Identification device of hiding article of hiding article, and the one, characteristics of human body is extracted and located, the 2nd, non-human target is identified.The process of wherein characteristics of human body being extracted and locate is mainly taking the aforesaid human body detecting method in mm-wave imaging and device as basis but be not limited to this human body detecting method and device.
Therefore,, as Figure 15, a kind of automatic detection and Identification method of hiding article comprises the following steps: S1, tested personnel are carried out to millimeter wave scanning obtain original image; S2, to described original image adjust obtain target image; S3, carry out cutting apart of human body and locate according to described target image; S5, generation bar rod combination model; S6, according to described original image, non-human target is detected, obtain non-human target distribution original image; S7, utilize described excellent combination model to obtain the location distribution information of described non-human target distribution original image with respect to human body; S8, described non-human target is carried out classification identification and shown that concealment article are with respect to the location distribution information of human body.
Correspondingly, a kind of automatic detection and Identification device of hiding article, comprising:
Scanister 1, for performing step S1, carries out millimeter wave scanning to tested personnel and obtains original image;
Adjusting module 2, for performing step S2, adjusts and obtains target image described original image;
Cut apart locating module 3, for performing step S3, carry out cutting apart of human body and locate according to described target image;
Bar rod combination model generation module, for performing step S5, generates bar rod combination model;
Non-human target Preliminary detection module, for performing step S6, detects non-human target according to described original image, obtains non-human target distribution original image;
Non-human target distribution module, for performing step S7, utilizes described excellent combination model to obtain the location distribution information of described non-human target distribution original image with respect to human body;
Classification recognition module, for performing step S8, carries out classification identification and shows the location distribution information of concealment article with respect to human body described non-human target.
Wherein step S1, S2 and S3 be in a kind of human body detecting method based on mm-wave imaging, scanister 1, adjusting module 2, cuts apart locating module 3 and describe in a kind of human body detection device based on mm-wave imaging, repeats no more herein.
According to Figure 16, step S5 further comprises following sub-step: S51, generates the bar-shaped model that the each key point of human body is provided; S52, generation provide the bar band model of human body contour outline information; S53, in conjunction with described bar-shaped model and described band model, thereby generate bar rod combination model.
Correspondingly, the bar rod combination model generation module of execution step S5 comprises following submodule:
Bar-shaped model generation module, for performing step S51, generates the bar-shaped model that the each key point of human body is provided;
Band model generation module, for performing step S52, generates the bar band model that human body contour outline information is provided;
Binding modules, for performing step S53, in conjunction with described bar-shaped model and described band model.
Wherein, the performed step S51 of bar-shaped model generation module is by a kind of step S32 of the human body detecting method based on mm-wave imaging, determines the coordinate of the each key point of described target image human body and obtain the step of the horizontal cut-off rule between partes corporis humani position, utilize the each key point of human body wherein to obtain the each articulation point that builds bar-shaped model, then connect each articulation point with straight line and just can generate bar-shaped model.
In the performed step S52 of band model generation module, generate the process of bar band model identical with the process that S4 in a kind of human body detecting method based on mm-wave imaging generates manikin.
The performed step S53 of binding modules is in conjunction with described bar-shaped model and described band model, generate bar rod combination model as shown in figure 17, thereby complete characteristics of human body's extraction and location in this method, articulation point when circular node in Figure 17 represents that characteristics of human body extracts, the sequence of extraction of each articulation point that can also add digitized representation in these articulation points.
Further, the non-human target Preliminary detection module of the step S6 of execution also comprises: rim detection module, highlight module, regularization module and Fusion Module.Correspondingly, as shown in figure 18, step S6, according to described original image, non-human target is detected, obtains non-human target distribution original image and also comprise following sub-step:
S61, rim detection module are carried out rim detection to described original image, for example, utilized Sobel (Sobel) operator to carry out rim detection, tentatively identify non-human target, as shown in figure 19;
S62, highlight module and highlight non-human target distribution region by mathematical morphological operation, such as, first Figure 19 is carried out to erosion operation, then carry out dilation operation.Wherein to use respectively the length of side be 2 and 4 square structure element to burn into dilation operation, thus the non-human target distribution areal map highlighting, as Figure 20;
S63, regularization module are chosen minimum circumscribed rectangle according to the border in described non-human target distribution region, make the irregular region in Figure 20 be converted into the non-human goal rule regional distribution chart in Figure 21;
S64, module merge Figure 21 and merge non-human goal rule regional distribution chart and described original image Fig. 3, obtain the non-human target original image of Figure 22, and this image has also demonstrated the human body parts of non-human target position certainly.
Because stick model only comprises human body contour outline information, there is no concrete articulation point information, bar-shaped model only comprises articulation point information, cover the position of non-human target on human body contour outline, so further, the performed step S7 of non-human target distribution module is by described non-human target distribution original image is input on described excellent combination model, so both can utilize the non-human target of human body contour outline information acquisition that bar rod combination model discal patch band model provides in the intramarginal distribution of human body contour outline, further utilize the human synovial dot position information of bar-shaped model in bar rod combination model to obtain the relative position relation between non-human target and articulation point, therefore utilize bar rod combination model, make non-human target on this object of reference of human body, there is location more accurately, the distribution plan of target original image as non-human in Figure 23 on bar rod combination model.
Further, classification recognition module comprises following submodule: exposed spots localization module, sort module and display module.Being respectively used to perform step S8 carries out class of risk identification to described non-human target and shows that concealment article are with respect to the each sub-steps in the location distribution information of human body, as Figure 25:
S81, exposed spots localization module position the exposed position of human body, as head, wrist, palm etc., can aforesaid human body cut apart and locate in carry out spots localization so more targetedly;
S82, sort module are classified to non-human target, the non-human target being distributed on the exposed position of human body are defined as to non-concealment article, as glasses, button, wrist-watch, ring etc.; Owing to there being clothing to hide, the non-human target outside the exposed position of human body cannot directly be checked by security staff, and the non-human target being distributed in outside the exposed position of human body is defined as hiding article, therefore needs to pay close attention to;
S83, display module are rejected the original image of described non-concealment article and are shown the distributed intelligence of concealment article original image on described excellent combination model, as Figure 24 has rejected the design sketch showing again after the original image of the wrist-watch of wrist in Figure 23.
By the automatic detection and Identification method and apparatus of concealment article of the present invention, can reduce personnel's request for utilization, reduce personal error, shorten the interpretation time that human body concealment dangerous material check.
On the other hand, after human body mm-wave imaging, because original image is more clear, by the automatic detection and Identification method of aforesaid human body detecting method based on mm-wave imaging and concealment article, the concealment article on human body and/or human body can be identified and show, but the exposure of privacy places of human body can be caused simultaneously.
In order to protect privacy, the present invention also provides a kind of method for secret protection based on mm-wave imaging, as Figure 26, comprising: S1, tested personnel are carried out to millimeter wave scanning obtain original image; A, according to described original image carry out human detection and concealment Articles detecting; B, determine the privacy places of human body; C, the privacy places of human body is shielded and indicates the concealment Item Information on human body.
Correspondingly, as shown in figure 27, the present invention also provides a kind of secret protection device based on mm-wave imaging, comprising:
Scanister 1, obtains original image for tested personnel being carried out to millimeter wave scanning;
Pick-up unit, for performing step A, carries out human detection and concealment Articles detecting according to described original image;
Privacy places determination module, for performing step B, determines the privacy places of human body;
Privacy mask module, for performing step C, shields and indicates the concealment Item Information on human body to the privacy places of human body.
Wherein steps A can be undertaken by the automatic detection and Identification method of aforesaid human body detecting method based on mm-wave imaging and concealment article, correspondingly, pick-up unit can comprise the human body detection device and the automatic detection and Identification device of hiding article based on mm-wave imaging.
The privacy places that the privacy places determination module of execution step B is determined human body equally according to aforementioned human body cut apart and location and human anatomy are determined privacy places and locate, comprise the sex that judges tested personnel, in the time that tested personnel are the male sex, human body head region and human body waist center are defined as to privacy places to the region of lower trunk width 1/2, in the time that tested personnel are women, be privacy places from trunk upper end down to the region of trunk height 1/2 to region and the trunk of lower trunk width 1/2 by human body head region and human body waist center.In the present invention, the mm-wave imaging taking the tested personnel of the male sex is described as example.
Execution step C, the privacy mask module that the privacy places of human body was shielded and indicated the concealment Item Information on human body can adopt following several embodiment to carry out the protection of privacy:
(1) privacy mask module comprises that part obfuscation module and first indicates module, part obfuscation module is carried out Fuzzy processing forming section obfuscation original image to privacy places on described original image, obfuscation can be used morphology operations, use the certain mosaic region of the length of side, or can directly use the rectangular block of solid color to cover privacy places; First indicates module marks described concealment article on described part obfuscation original image with sign frame, such as the frame of highlight color, as shown in figure 28.This mode is adapted to hide the situation that article detect automatically and manual detection combines.Or,
(2) privacy mask module comprises that selection object image module and second indicates module, because the image of the binaryzation of complete display can mask the Pixel Information of tested personnel's privacy places, so present embodiment utilization selects object image module to select object image representation human body parts, this object image obtains by the described human body detecting method based on mm-wave imaging, as Fig. 8; Show concealment Item Information on this object image before, second indicates module first judges the whether privacy places in human body of concealment article, if, use the color blocks different from human body color represent to hide article and indicate on described object image, as use certain gray-scale value, as 128 grey rectangle piece indicates; If not, directly the original image of concealment article is presented on described object image.Shielded the complete information of human body in original image so completely, reached again the object of secret protection, effect as shown in figure 29.This method is only applicable to hide under the automatic detection mode of article, does not need to show the situation of whole person's volume image.Or,
(3) privacy mask module comprises that whole obfuscation modules and the 3rd indicate module.All obfuscation module is carried out whole Fuzzy processing to human body and is formed whole obfuscation original images on described original image, and the method for obfuscation can be identical with embodiment (1), and in present embodiment, obfuscation has been used operator
Before showing concealment Item Information on these whole obfuscation original images, the 3rd indicates module judge and hides the whether privacy places in human body of article, if, use the color blocks different from human body color represent to hide article and indicate on described whole obfuscation original images, as use certain gray-scale value, as 128 grey rectangle piece indicates; If not, the original image of concealment article is presented on described whole obfuscation original image, design sketch is as Figure 30.This method is only applicable to hide under the automatic detection mode of article, does not need to show the situation of whole person's volume image.Or,
(4) privacy mask module comprises that preference pattern module and the 4th indicates module.Preference pattern module is selected the manikin in described human detection, and the manikin here can be bar band model, can be also bar rod combination model; The 4th indicates module uses the color blocks different from manikin color, as redness, represents concealment article and indicates on described manikin.So also can shield the privacy information of human body completely.The effect showing on bar band model as shown in figure 31.This method is only applicable to not show that the complete information of hiding article does not need to show the situation of whole person's volume image yet.
Should be appreciated that the above detailed description of technical scheme of the present invention being carried out by preferred embodiment is illustrative and not restrictive.Those of ordinary skill in the art modifies reading the technical scheme that can record each embodiment on the basis of instructions of the present invention, or part technical characterictic is wherein equal to replacement; 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.Protection scope of the present invention is only limited by the claims of enclosing.
Claims (8)
1. the human body detection device based on mm-wave imaging, is characterized in that, comprising:
Scanister, obtains original image for tested personnel being carried out to millimeter wave scanning;
Adjusting module, obtains target image for described original image is adjusted;
Cut apart locating module, for carrying out cutting apart of human body and locate according to described target image, described in cut apart locating module and comprise
Vertically center line module, for determining the vertical center line of human body;
Coordinate horizontal line module, for the vertical center line based on human body, determines the coordinate of the each key point of described target image human body and obtains the horizontal cut-off rule between partes corporis humani position;
Width slope module, for the horizontal cut-off rule based between partes corporis humani position, determines width and the slope of partes corporis humani position;
Manikin generation module, for cutting apart and locating, generation manikin based on human body.
2. the human body detection device based on mm-wave imaging according to claim 1, it is characterized in that, described manikin generation module is further used for according to width and the slope of the coordinate of described each key point, described partes corporis humani position, obtains the manikin with rectangle and/or parallelogram composition.
3. the human body detection device based on mm-wave imaging according to claim 1, is characterized in that, described adjusting module also comprises following submodule:
Pretreatment module, obtains preliminary denoising image for described original image being carried out to pre-service;
Binarization block, obtains preliminary bianry image for described preliminary denoising image is carried out to binaryzation;
Processing module again, obtains described target image for described preliminary bianry image is processed again.
4. the human body detection device based on mm-wave imaging according to claim 3, is characterized in that, described pretreatment module further comprises with lower unit:
Difference arithmetic element, for carrying out difference computing by described original image and background image gray-scale value;
Smoothing processing unit, for carrying out picture smooth treatment;
Linear change unit, for carrying out linear greyscale transformation.
5. the human body detection device based on mm-wave imaging according to claim 3, is characterized in that, described binarization block further utilizes Pulse Coupled Neural Network algorithm to be criterion to the maximum and to choose the threshold value of binaryzation with entropy.
6. the human body detection device based on mm-wave imaging according to claim 3, is characterized in that,
Described processing module is more further processed by morphologic filtering again.
7. the human body detection device based on mm-wave imaging according to claim 6, is characterized in that, describedly comprises by morphologic filtering:
Use the square structure element that the length of side is 5 to carry out the bright noise spot outside erosion operation elimination human body;
The square structure element that the use length of side is 4 carries out eliminating when opening operation keeps image size isolated area and the burr at human body edge;
When carrying out closed operation maintenance image size, the square structure element that the use length of side is 4 fills the tiny cavity in human body and the edge of level and smooth human body;
The square structure element that the use length of side is 5 carries out dilation operation makes image return to life size.
8. the human body detection device based on mm-wave imaging according to claim 3, is characterized in that, described pretreatment module also comprises:
Image enhancing unit, for carrying out figure image intensifying to described original image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210050292.0A CN102707322B (en) | 2012-02-29 | 2012-02-29 | Human body detection device based on millimeter wave imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210050292.0A CN102707322B (en) | 2012-02-29 | 2012-02-29 | Human body detection device based on millimeter wave imaging |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102707322A CN102707322A (en) | 2012-10-03 |
| CN102707322B true CN102707322B (en) | 2014-11-12 |
Family
ID=46900251
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210050292.0A Active CN102707322B (en) | 2012-02-29 | 2012-02-29 | Human body detection device based on millimeter wave imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102707322B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3471058A4 (en) * | 2016-06-13 | 2020-01-15 | Shenzen Cct Thz Technology Co., Ltd. | Method and apparatus for detecting human body gender in microwave image |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102013016813B4 (en) * | 2012-12-19 | 2023-09-07 | Sony Corporation | Method of operating a portable screening device and portable screening device |
| CN103903299B (en) * | 2012-12-27 | 2018-05-22 | 同方威视技术股份有限公司 | The three-dimensional Enhancement Method and equipment of back scattering human body check image |
| CN105513035B (en) * | 2014-09-22 | 2018-12-04 | 北京计算机技术及应用研究所 | The detection method and its system of human body concealment article in a kind of passive millimeter wave image |
| CN106447634B (en) * | 2016-09-27 | 2018-02-27 | 中国科学院上海微系统与信息技术研究所 | A kind of privacy places positioning and guard method based on active mm-wave imaging |
| CN106485620A (en) * | 2016-10-11 | 2017-03-08 | 深圳万发创新进出口贸易有限公司 | A kind of Dormitory management system based on cloud computing |
| CN106485077B (en) * | 2016-10-11 | 2018-12-14 | 徐州玖胜医疗器械有限公司 | A kind of supervision place medical monitoring systems |
| CN106291738B (en) * | 2016-10-11 | 2018-06-29 | 江苏科力斯通新材料有限公司 | A kind of safe examination system on bus |
| CN106272479A (en) * | 2016-10-11 | 2017-01-04 | 深圳万发创新进出口贸易有限公司 | A kind of intelligence reconnaissance robot |
| CN106959305A (en) * | 2017-03-25 | 2017-07-18 | 合肥赛英迪光电科技有限公司 | A kind of contraband detecting instrument and contraband detecting method |
| CN108120978A (en) * | 2017-12-13 | 2018-06-05 | 北京无线电计量测试研究所 | Human body attitude precise recognition method in a kind of mm-wave imaging |
| CN109031456A (en) * | 2018-05-30 | 2018-12-18 | 深圳市华讯方舟太赫兹科技有限公司 | A kind of safe examination system and safety inspection method |
| CN109597069A (en) * | 2018-12-25 | 2019-04-09 | 山东雷诚电子科技有限公司 | A kind of active MMW imaging method for secret protection |
| CN112099001B (en) * | 2020-09-18 | 2021-09-03 | 欧必翼太赫兹科技(北京)有限公司 | Control method of three-dimensional special-shaped planar aperture holographic imaging security inspection radar |
| CN112581391B (en) * | 2020-12-14 | 2023-04-11 | 佛山科学技术学院 | Image denoising method based on PCNN |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006071922A2 (en) * | 2004-12-29 | 2006-07-06 | Siemens Medical Solutions Usa, Inc. | Combined pet/mr imaging system and apd-bassed pet detector for use in simultaneous pet/mr imaging |
| WO2010092368A2 (en) * | 2009-02-10 | 2010-08-19 | Durham Scientific Crystals Limited | Apparatus and method for viewing an object |
| CN102135610A (en) * | 2010-12-28 | 2011-07-27 | 中国航天科工集团第二研究院二○三所 | Near-field real-time calibration method for human body millimeter wave imaging safety inspection system |
-
2012
- 2012-02-29 CN CN201210050292.0A patent/CN102707322B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006071922A2 (en) * | 2004-12-29 | 2006-07-06 | Siemens Medical Solutions Usa, Inc. | Combined pet/mr imaging system and apd-bassed pet detector for use in simultaneous pet/mr imaging |
| WO2010092368A2 (en) * | 2009-02-10 | 2010-08-19 | Durham Scientific Crystals Limited | Apparatus and method for viewing an object |
| CN102135610A (en) * | 2010-12-28 | 2011-07-27 | 中国航天科工集团第二研究院二○三所 | Near-field real-time calibration method for human body millimeter wave imaging safety inspection system |
Non-Patent Citations (1)
| Title |
|---|
| 权利要求1到权利要19,附图1到附图4，. * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3471058A4 (en) * | 2016-06-13 | 2020-01-15 | Shenzen Cct Thz Technology Co., Ltd. | Method and apparatus for detecting human body gender in microwave image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102707322A (en) | 2012-10-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102707322B (en) | Human body detection device based on millimeter wave imaging | |
| CN102697503B (en) | Human detection method based on millimeter wave imaging | |
| CN102629315B (en) | A kind of detection automatically hiding article and identification device | |
| CN102708372B (en) | A kind of automatic detection and Identification method of hiding article | |
| CN102708560B (en) | A kind of method for secret protection based on mm-wave imaging | |
| CN102708561B (en) | A kind of secret protection device based on mm-wave imaging | |
| CN105469113B (en) | A kind of skeleton point tracking method and system in two-dimensional video stream | |
| US7253766B2 (en) | Three-dimensional surface/contour processing based on electromagnetic radiation interrogation | |
| CN109447071A (en) | A kind of mm-wave imaging dangerous goods detection method based on FPGA and deep learning | |
| CN108665487A (en) | Substation's manipulating object and object localization method based on the fusion of infrared and visible light | |
| US20090174554A1 (en) | Method and system for screening luggage items, cargo containers or persons | |
| CN110186940B (en) | Security check identification method and device, computer equipment and storage medium | |
| CN104599273B (en) | Sea and sky background infrared small target detection method based on multi-scale wavelet crossing operation | |
| CN109492714A (en) | Image processing apparatus and its method | |
| CN102567733B (en) | Detection and shielding method for private parts of human bodies under non-cooperation condition in microwave images | |
| CN106447634B (en) | A kind of privacy places positioning and guard method based on active mm-wave imaging | |
| CN109544563A (en) | A kind of passive millimeter wave image human body target dividing method towards violated object safety check | |
| CN106991417A (en) | A kind of visual projection's interactive system and exchange method based on pattern-recognition | |
| WO2018035814A1 (en) | Millimetre wave image-based human body foreign matter detection method and system | |
| CN106372583A (en) | Millimeter wave image-based human body foreign matter detection method and system | |
| Xu et al. | YOLO-MSFG: toward real-time detection of concealed objects in passive terahertz images | |
| CN110378924A (en) | Level set image segmentation method based on local entropy | |
| CN106326834A (en) | Human body gender automatic identification method and apparatus | |
| Yang et al. | A novel deformable body partition model for MMW suspicious object detection and dynamic tracking | |
| US10445591B2 (en) | Automated target recognition based body scanner using database scans |
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 |