WO2019047492A1 - 一种人体图像映射方法、系统及终端设备 - Google Patents
一种人体图像映射方法、系统及终端设备 Download PDFInfo
- Publication number
- WO2019047492A1 WO2019047492A1 PCT/CN2018/078038 CN2018078038W WO2019047492A1 WO 2019047492 A1 WO2019047492 A1 WO 2019047492A1 CN 2018078038 W CN2018078038 W CN 2018078038W WO 2019047492 A1 WO2019047492 A1 WO 2019047492A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- human body
- image
- ordinate
- spatial distribution
- distribution histogram
- Prior art date
Links
- 238000013507 mapping Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001514 detection method Methods 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 22
- 238000010276 construction Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 5
- 210000003414 extremity Anatomy 0.000 description 92
- 210000000038 chest Anatomy 0.000 description 33
- 210000003423 ankle Anatomy 0.000 description 22
- 210000003127 knee Anatomy 0.000 description 21
- 230000000875 corresponding effect Effects 0.000 description 20
- 210000002683 foot Anatomy 0.000 description 16
- 238000003384 imaging method Methods 0.000 description 15
- 210000001015 abdomen Anatomy 0.000 description 14
- 230000006870 function Effects 0.000 description 12
- 239000002184 metal Substances 0.000 description 8
- 229910052751 metal Inorganic materials 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000000877 morphologic effect Effects 0.000 description 6
- 238000005530 etching Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 230000003187 abdominal effect Effects 0.000 description 3
- 210000000481 breast Anatomy 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 3
- 238000005260 corrosion Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 210000003128 head Anatomy 0.000 description 2
- 230000036544 posture Effects 0.000 description 2
- 210000000689 upper leg Anatomy 0.000 description 2
- 210000000746 body region Anatomy 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000013043 chemical agent Substances 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- OWZREIFADZCYQD-NSHGMRRFSA-N deltamethrin Chemical compound CC1(C)[C@@H](C=C(Br)Br)[C@H]1C(=O)O[C@H](C#N)C1=CC=CC(OC=2C=CC=CC=2)=C1 OWZREIFADZCYQD-NSHGMRRFSA-N 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 239000010977 jade Substances 0.000 description 1
- 210000000629 knee joint Anatomy 0.000 description 1
- 210000002414 leg Anatomy 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052755 nonmetal Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 210000002832 shoulder Anatomy 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/04—Context-preserving transformations, e.g. by using an importance map
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Definitions
- the embodiments of the present invention belong to the field of image processing technologies, and in particular, to a human body image mapping method, system, and terminal device.
- the existing millimeter wave imaging security devices usually directly acquire the whole body image of the human body. Since the whole body image of the human body includes the image of the privacy part of the human body, the privacy of the user may be leaked, causing certain adverse effects.
- Embodiments of the present invention provide a human body image mapping method, system, and terminal device, which are intended to solve the current conventional millimeter wave security inspection equipment, which usually directly acquires a whole body image of a human body, because the whole body image of the human body includes a human body's privacy part. Images, which may cause the user's privacy to be leaked, cause certain adverse effects.
- a first aspect of the embodiments of the present invention provides a human body image mapping method, where the method includes:
- a second aspect of the present invention provides a human body image mapping system, the system comprising:
- a human body grayscale image acquiring unit for acquiring a millimeter wave grayscale image of the human body
- a human body contour image extracting unit configured to extract a human body contour image in the millimeter wave grayscale image
- a histogram construction unit configured to construct a vertical spatial distribution histogram of the human body contour image in a vertical direction and a horizontal spatial distribution histogram in a horizontal direction;
- a first limb position acquiring unit configured to acquire a limb position of the human body in the human body contour image according to the vertical spatial distribution histogram, the horizontal spatial distribution histogram, and a preset human body scale model;
- a second limb position acquiring unit configured to acquire a limb position of the human body in the cartoon image of the human body
- a mapping unit configured to construct a position mapping relationship between the human body contour image and each limb in the human body cartoon image according to a limb position of the human body in the human body contour image and a limb position of the human body in the human body cartoon image.
- a third aspect of the embodiments of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and operable on the processor, when the processor executes the computer program The steps of the above method.
- a fourth aspect of an embodiment of the present invention provides a computer readable storage medium storing a computer program, the computer program being executed by a processor to implement the steps of the above method.
- the embodiment of the invention extracts the human body contour image in the millimeter wave gray image by acquiring the millimeter wave gray image of the human body, and distributes the vertical space distribution histogram in the vertical direction and the horizontal spatial distribution in the horizontal direction according to the human body contour image.
- the histogram acquires the position of the human body in the contour image of the human body, acquires the position of the human body in the cartoon image of the human body, and constructs a positional mapping relationship between the contour image of the human body and the limbs in the human cartoon image, which can realize the contour image of the human body to the human body
- the mapping of cartoon images protects user privacy.
- Embodiment 1 is a basic flow chart of a human body image mapping method according to Embodiment 1 of the present invention
- Embodiment 2 is a millimeter wave grayscale image provided by Embodiment 1 of the present invention.
- Embodiment 3 is a binarized image provided by Embodiment 1 of the present invention.
- Embodiment 4 is a human body contour image according to Embodiment 1 of the present invention.
- FIG. 5 is a vertical spatial distribution histogram according to Embodiment 1 of the present invention.
- Embodiment 6 is a horizontal space distribution histogram provided by Embodiment 1 of the present invention.
- FIG. 7 is a schematic diagram showing a mapping relationship between a human body contour image and a human body cartoon image according to Embodiment 1 of the present invention.
- FIG. 8 is a schematic diagram showing the result of detecting foreign matter according to Embodiment 1 of the present invention.
- FIG. 9 is a basic flow chart of a human body image mapping method according to Embodiment 2 of the present invention.
- FIG. 10 is a structural block diagram of a human body image mapping system according to Embodiment 4 of the present invention.
- FIG. 11 is a schematic structural diagram of a first limb position acquiring unit according to Embodiment 5 of the present invention.
- FIG. 12 is a schematic structural diagram of a terminal device according to Embodiment 7 of the present invention.
- Embodiment 1 is a diagrammatic representation of Embodiment 1:
- this embodiment provides a human body image mapping method, including:
- Step S110 Acquire a millimeter wave grayscale image of the human body.
- the human body can be lifted up by the top of the head or the hands are raised to the position of the same height as the shoulder, or the human hands can be naturally sagged, or other standing postures conforming to the security standards can be used.
- the embodiment does not specifically limit the standing posture of the human body, and then uses millimeter wave data acquisition devices (for example, millimeter wave transceivers) to acquire millimeter wave data of the front or back of the human body, and utilizes a millimeter wave imaging system (for example, a millimeter wave imager).
- the millimeter wave data of the human body is processed into a millimeter wave grayscale image of the front or back of the human body.
- Step S120 Extract a human body contour image in the millimeter wave grayscale image.
- the human body contour image refers to an image corresponding to the shape of the limb contour of the human body.
- step S120 specifically includes:
- Step S121 Perform gray scale segmentation on the millimeter wave grayscale image I(x, y) to obtain a binarized image B(x, y) of the millimeter wave grayscale image I(x, y).
- FIG. 3 exemplarily shows a schematic diagram of a binarized image B(x, y).
- step S121 grayscale segmentation of the millimeter wave image gradation I(x, y) is performed, and obtaining a corresponding binarized image B(x, y) can be realized according to the following formula:
- x is the number of image columns
- y is the number of image rows
- X is the maximum number of columns in the image
- Y is the maximum number of rows in the image
- T is the threshold
- the value of 255 in B(x, y) represents the human body region.
- Step S122 performing morphological operations on the binarized image B(x, y) in both the horizontal direction and the vertical direction to generate a human body contour image B1(x, y), such as the human body contour image B1(x, y) Figure 4 shows.
- the horizontal direction and the vertical direction in step S122 refer to directions perpendicular and parallel to the human body height direction, respectively.
- step S122 specifically includes:
- a morphological expansion operation and a corrosion operation are performed on the B(x, y) image in the horizontal direction and the vertical direction, the kernel function of the expansion operation is 1 ⁇ 3, and the kernel function of the etching operation is 3 ⁇ 1
- the size, the expression of the expansion operation and the corrosion operation are respectively:
- x' and y' represent the translation unit values corresponding to the kernel function.
- the binarized image B(x, y) when performing the above morphological operation, may be subjected to an expansion operation and then subjected to an etching operation to obtain a human body contour image B1(x, y);
- the valued image B(x, y) is subjected to an squeezing operation and then subjected to an expansion operation to obtain a human body contour image B1(x, y).
- the present invention preferably performs an etching operation on the binarized image B(x, y) and then performs an etching operation to obtain a contoured human contour image B1 (x). , y).
- Step S130 Construct a vertical spatial distribution histogram of the human body contour image in the vertical direction and a horizontal spatial distribution histogram in the horizontal direction.
- the spatial distribution histogram in this embodiment refers to a histogram obtained by counting the gray value information of the human body contour image according to the spatial position.
- the vertical spatial distribution histogram specifically refers to the spatial position from left to right (from left to right, when the human body stands on the horizontal plane, from the left side of the human body to the right side and parallel to the horizontal plane) as the abscissa, the human body
- the frequency at which the gray value of the contour image appears within the length of the unit space position is a histogram of the ordinate.
- the horizontal spatial distribution histogram specifically refers to the spatial position from top to bottom (from top to bottom, when the human body stands on a horizontal plane, from the top of the human body to the sole of the foot and perpendicular to the horizontal plane) as the ordinate.
- the frequency at which the gray value of the human body contour image appears within the length of the unit space position is a histogram of the abscissa.
- the vertical spatial distribution histogram H is as shown in FIG. 5, and in step S130, a vertical space distribution histogram H of the human body contour image B1(x, y) in the vertical direction is constructed, and the following formula can be adopted. achieve:
- the vertical space distribution histogram H is smoothed, and the formula for setting the smoothing scale to 3 is as follows:
- the horizontal spatial distribution histogram V is as shown in FIG. 6.
- a horizontal spatial distribution histogram V of the human body contour image B1(x, y) in the horizontal direction is constructed, which can be determined by the following formula. achieve:
- the vertical space distribution histogram V is smoothed, and the formula for setting the smoothing scale to 3 is as follows:
- Step S140 Acquire a limb position of the human body in the human body contour image according to the vertical spatial distribution histogram, the horizontal spatial distribution histogram, and the preset human body scale model.
- the human body scale model is constructed according to the body structure of the human body, the size ratio between the limbs, and the shape of the limb.
- the position of the limb specifically includes the head, shoulders, chest, abdomen, ankle, limbs (both hands, feet) and knee joint position of the human body, and in special cases, for the disabled person, the limb position Corresponding to the missing part of the data.
- Step S150 Acquire a limb position of the human body in the cartoon image of the human body.
- the human body cartoon image may be a humanoid image simulated by a computer in advance, and the position coordinates of each limb in the image may be set in advance, that is, the human body position of the human body cartoon image is It is known data that can be set in advance by the staff. When it is necessary to use the known data, it is only necessary to directly call the data of the body position of the human body in the human body cartoon image set and stored in advance. In other applications, the position of the human body in the cartoon image of the human body can also be obtained in the same manner as the position of the body of the human body in the image of the human body contour. Specifically, the following steps can be implemented:
- Step S160 Construct a position mapping relationship between the human body contour image and each limb in the human body cartoon image according to the limb position of the human body in the human body contour image and the limb position of the human body in the human body cartoon image.
- the positional mapping relationship of each limb refers to the mapping relationship of the position coordinates of each limb.
- the positional mapping relationship of the head is the coordinates (A, B).
- the head position coordinates in the above examples are merely exemplary, since in practice, each limb is displayed in the form of a two-dimensional planar image in the image, so the position corresponding to each limb There may be more than one coordinate.
- the position coordinate of the geometric center of the limb image may be used as the position coordinate of the limb, or two or more coordinate points may be selected on the limb map, and the position coordinates of the selected coordinate point may be selected. As the position coordinates of the limb.
- step S160 before step S160,
- the two images before establishing the mapping relationship between the precise position coordinates of each limb in the two images, the two images can be divided into equal numbers of image regions by the same image region division manner, by establishing two images in the image.
- the mapping relationship between image regions can roughly achieve the correspondence of limb positions in the two images.
- the position coordinates of each limb may be further correlated by step S160 based on the mapping relationship of the image regions.
- a more precise positional correspondence can be achieved by further finely dividing the two images, which can be achieved by increasing the number of the first image area and the second image area.
- FIG. 7 a schematic diagram showing when the human body contour image and the human body cartoon image are divided into nine image regions is exemplarily shown, and the image regions having the same label in the figure have a mapping relationship. Also shown in Fig. 7 is a coordinate point corresponding to two position coordinates, denoted as point A and point a, respectively, showing the distance between the two coordinate points and the boundary of their respective image areas 5.
- the mapping relationship between two coordinate points can be obtained; similarly, the mapping relationship between the human contour image and any two points in the human cartoon image can be obtained.
- the human body image mapping method further includes:
- Step S170 Identify foreign objects in the contour image of the human body according to the preset foreign object feature recognition model.
- the foreign matter may be a metal-based foreign matter such as a metal gun, a cutter, or a gold bullion, or may be a non-metallic foreign matter such as a chemical agent, an ivory, or a jade.
- step S170 includes:
- a position of a foreign object at an edge of the human contour image is determined according to a well-defined area of the edge of the human contour image in the millimeter wave grayscale image.
- the above-described foreign matter identification method can greatly improve the accuracy of foreign matter detection and can recognize metal foreign matter and non-metal foreign matter.
- Step S180 Identify the position of the foreign object on the human body contour image according to the limb position of the human body in the human body contour image.
- the position of the foreign object on the contour image of the human body is specifically determined, that is, the position of the contour image of the human body corresponding to each limb is identified, and the foreign object is identified, for example, when the foreign object is in the abdomen of the human body, the contour image of the human body is displayed.
- the location of the abdomen identifies the foreign object.
- Step S190 Identify, according to the location mapping relationship, a location of the foreign object on the cartoon image of the human body and generate a foreign object detection result.
- the method of identifying the foreign matter in steps S190 and S180 is the same, and the only difference is that the human body contour image is replaced with the human body cartoon image.
- the position of the foreign object is directly marked with a distinct rectangular frame at the position of the abdomen in the contour image of the human body and the position directly in front of the thigh.
- the foreign matter detection result may be presented in the form of characters, images, voices, tables, etc., for example, when there is a foreign body on the chest of the human body, the words “chest” or more specifically “metal foreign bodies in the chest” may be presented in the text.
- the human body contour image in the millimeter wave grayscale image is extracted by acquiring the millimeter wave grayscale image of the human body, and the vertical spatial distribution histogram in the vertical direction and the horizontal spatial distribution histogram in the horizontal direction according to the human body contour image.
- the figure obtains the position of the human body in the contour image of the human body, acquires the position of the human body in the cartoon image of the human body, and constructs the positional mapping relationship between the contour image of the human body and the limbs in the cartoon image of the human body, and can realize the contour image of the human body to the human body cartoon Image mapping to protect user privacy.
- Embodiment 2 is a diagrammatic representation of Embodiment 1:
- step S140 includes:
- Step S141 Perform local maximum detection on the vertical spatial distribution histogram to obtain a vertical midline abscissa of the human body.
- step S141 specifically includes:
- Step S1411 Extract the upper half of the vertical distribution histogram H as a sub-vertical distribution histogram H 1 .
- the sub-vertical distribution histogram H 1 is the upper half of the vertical distribution histogram H, that is, extracted according to 50%*H.
- the upper half of the vertical distribution histogram H can also be extracted according to other ratios. In part, the extraction ratio is not limited in this embodiment.
- Step S1412 the maximum value of the histogram acquisition sub-vertical distribution of H 1.
- the formula for obtaining this maximum value is as follows:
- the column of x mid is the position where the abscissa of the vertical center line is located, and the line perpendicular to the horizontal lines passing through the head, chest, abdomen and ankle of the human body as shown in FIG. 4 is It is an even vertical center line.
- obtaining the vertical centerline abscissa according to the vertical distribution histogram may also be implemented by other methods, which is not illustrated herein.
- Step S142 Perform local minimum detection on the horizontal spatial distribution histogram, and acquire the human hand's top ordinate, the top ordinate, and the sole ordinate.
- step S142 specifically includes:
- Step S1421 Perform local minimum detection on the first region of the horizontal spatial distribution histogram, and determine the vertical coordinate of the human hand according to the detected local minimum value.
- step S1421 is specifically:
- Performing local minimum detection on the first region of the horizontal spatial distribution histogram, and detecting the local minimum is the human hand top ordinate y hand ; wherein the first region is a closed interval [vertical midline abscissa x mid - first set value, vertical midline abscissa x mid + second set value].
- the first set value and the second set value may be the same or different, and may be set according to actual needs of the user.
- the size of the first area is defined as 101 pixels, and the first set value and the second set value are the same, and the first partial area is [x mid -50, x mid +50].
- Step S1422 Acquire a number of pixel points in each row of the second area of the horizontal spatial distribution histogram as a first preset pixel threshold, according to a row in which the number of pixels is smaller than a first preset threshold. Determine the ordinate of the head of the human body.
- step S1422 is specifically:
- the coordinates of the row of the line are determined as the y head of the human body, wherein the second region is a closed interval [vertical midline abscissa x mid - third set value, vertical midline abscissa x mid + fourth Set value].
- the third set value and the fourth set value may be the same or different, and may be set according to actual needs of the user.
- the size of the second area is defined as 31 pixels, and the third set value and the fourth set value are the same, and the second partial area is [x mid -15, x mid +15].
- Step S1423 Acquire a number of pixel points in each row of the third region of the horizontal spatial distribution histogram as a second preset pixel threshold, according to a row in which the number of pixels is smaller than a second preset threshold. Determine the ordinate of the sole of the human body.
- step S1423 is specifically:
- the coordinates of the row of the line are determined as the sole y foot of the human body, wherein the third region is a closed interval [between the bottom of the horizontal spatial distribution histogram and the bottom of the horizontal spatial distribution histogram The distance is the position of the fifth set value].
- the fifth set value may be set according to actual needs.
- the size of the third region is defined as 60 pixels, and the horizontal space distributes the area between the bottom of the histogram to a position 60 pixels from the bottom.
- the second preset pixel threshold may be set to 255, and when obtaining the number of pixel points whose pixel value is less than 255 in each row, each row of images may be detected from top to bottom in the third region.
- the pixel value of the pixel is 255, and the number of pixels in each row is 255 pixels.
- the number of pixels is less than the preset threshold, that is, the position of the sole y foot .
- Step S143 Acquire the height of the human body according to the top ordinate of the head and the ordinate of the sole.
- the formula for obtaining the height height of the human body in step S143 is:
- Step S144 Obtaining according to the overhead ordinate or the sole ordinate and the vertical midline abscissa, the height, the vertical spatial distribution histogram, the horizontal spatial distribution histogram, and the preset human body scale model. A contour image of the limb corresponding to each limb of the human body.
- step S144 includes:
- Step S1441 Obtain a shoulder of the human body according to any one of the overhead ordinate or the sole ordinate and the vertical midline abscissa, the height, the horizontal spatial distribution histogram, and a preset human body scale model. An ordinate and a width of a head of the human body relative to the abscissa of the vertical centerline;
- Step S1442 Acquire an accurate width of the head according to the height, the body part proportional probability model, and the vertical spatial distribution histogram, and acquire a head contour image;
- Step S1443 According to the image information above the shoulder in the human body contour image, the head contour image is removed, and the left and right arms contour images of the human body are obtained.
- an edge detection operation may be performed on the left and right arms contour images to obtain edge contours of the left and right arms of the human body, and the left and right arms may be obtained according to the shape characteristics of the arms of the human body and the curvature variation law of the contours.
- the coordinates of the elbow portion are cut, and the inflection point positions of the left and right elbow portions are cut, and the arm contour image and the arm contour image of the left and right arms are respectively obtained.
- the shoulder coordinates of the human body can be obtained according to the overhead ordinate or the sole ordinate and the height, the horizontal spatial distribution histogram, and the preset human body scale model.
- Step S1441 specifically includes:
- Step S1441-1 Obtaining the shoulder of the human body according to the top ordinate y head or the sole y foot and the height height, the height of the human body and the chest ratio HEIGHT_CHEST_RATIO in the preset human body scale model
- the first ordinate y shoulder When calculated using the overhead ordinate y head , the expression of the shoulder first ordinate y shoulder is as follows:
- y shoulder height*HEIGHT_CHEST_RATIO+y head ;
- HEIGHT_CHEST_RATIO is an average value obtained based on the measured values of the height of the human body and the horizontal position of the shoulder of the human body in the N-dimensional millimeter-wave grayscale image. In the present embodiment, it is preferable that HEIGHT_CHEST_RATIO is 0.2.
- the shoulder first ordinate y shoulder is the approximate value of the shoulder ordinate obtained.
- the sole ordinate y foot can also be used to determine the first ordinate of the shoulder of the human body, which is not limited in this embodiment.
- Step S1441-2 Obtain a peak region near the first longitudinal coordinate of the shoulder according to the horizontal spatial distribution histogram, and determine a vertical coordinate corresponding to the peak region as a shoulder second coordinate, wherein the peak region The distance from the first ordinate of the chest is less than the first predetermined distance threshold.
- Step S1441-3 According to the imaging characteristics of the shoulder of the human body, the horizontal spatial distribution histogram near the shoulder ordinate position has a peak region, and the corresponding ordinate is y' shoulder , that is, the second shoulder of the human body coordinate.
- Step S1441-4 obtaining the shoulder ordinate according to the shoulder first ordinate and the shoulder second ordinate
- the expression is as follows:
- Step S1443 includes:
- S1443-1 According to the gray-scale image of the left and right arms of the human body, the discontinuous plaque is formed and the imaging area of the human tissue is large, and the plaque area with a larger imaging area is obtained as the area where the two-arm contour image is located;
- S1443-2 calculating a ratio of the first imaging area and the second imaging area in the plaque area to the area of the plaque area, if the ratio is within a preset interval, the first imaging area and the The second imaging area is used as the dual-arm contour image. If the ratio is not within the preset interval, it indicates that the first imaging area includes the dual-arm contour image, and the external image is fitted according to the first imaging area. The curvature change of the polygon is segmented to obtain the boom contour image and the arm contour image.
- the preset interval is preferably [1, 10].
- step S144 further includes:
- Step S1444 includes:
- Step S1444-1 obtaining a first vertical coordinate y of the human chest according to the top ordinate y head or the human sole ordinate y foot and the height height, the height and the chest ratio HEIGHT_CHEST_RATIO in the preset human body scale model. Chest .
- the expression of the first ordinate of the chest is as follows:
- y chest height*HEIGHT_CHEST_RATIO+y head ;
- HEIGHT_CHEST_RATIO is an average value obtained based on the measured values of the height of the human body and the horizontal position of the chest of the human body in the millimeter-wave grayscale image of the N human body, and is preferably 0.8 in the present embodiment.
- the first ordinate y chest of the chest is the approximate value of the ordinate of the chest.
- the sole ordinate y foot can also be used to determine the first ordinate of the chest, which is not limited in this embodiment.
- Step S1444-2 obtaining a peak region near the first longitudinal axis of the chest according to the horizontal spatial distribution histogram, and determining a vertical coordinate corresponding to the peak region as a second longitudinal coordinate of the chest, wherein the peak region and the The distance between the first ordinate of the chest is less than the second preset distance threshold;
- Step S1444-3 According to the imaging characteristics of the contour image of the human chest, the horizontal spatial distribution histogram near the position of the first ordinate of the chest has a peak region, and the corresponding ordinate is y' chest , which is The second ordinate of the chest.
- Step S1444-4 obtaining the chest ordinate according to the first ordinate of the chest and the second ordinate of the chest
- the expression is as follows:
- step S144 further includes:
- Step S1445 Acquire the abdominal longitudinal coordinate of the human body.
- Step S1445 specifically includes:
- the first longitudinal coordinate y abdomen of the human body chest is obtained according to the top ordinate y head or the human sole ordinate y foot and the height height, the ratio of the height and the abdomen HEIGHT_ABDOMEN_RATIO in the preset human body scale model.
- the expression of the first ordinate of the abdomen is as follows:
- y abdomen height*HEIGHT_ABDOMEN_RATIO+y head ;
- HEIGHT_ABDOMEN_RATIO is an average value obtained based on the measured values of the height of the human body and the horizontal position of the abdomen of the human body in the N-dimensional human millimeter wave grayscale image, and is preferably 0.44 in this example.
- the first vertical coordinate y abdomen of the human abdomen is the approximate value of the obtained abdominal longitudinal coordinate.
- the sole ordinate y foot can also be used to determine the first ordinate of the abdomen, which is not limited in this embodiment.
- step S144 further includes:
- Step S1446 Acquire the longitudinal coordinate of the ankle of the human body.
- Step S1446 specifically includes:
- Step S1446-1 obtaining the first longitudinal direction of the human body according to the top ordinate y head or the human sole ordinate y foot and the height height, the ratio of the height and the ankle height HEIGHT_CROTCH_RATIO in the preset human body scale model. Coordinate y crotch .
- the expression of the first ordinate y crotch of the ankle is as follows:
- y crotch height*HEIGHT_CROTCH_RATIO+y head ;
- HEIGHT_CROTCH_RATIO is an average value obtained based on the measured values of the height of the human body and the horizontal position of the crotch portion of the human body in the millimeter-wave grayscale image of the human body, and is preferably 0.51 in the present embodiment.
- the first ordinate y crotch of the ankle is the approximate value of the ordinate of the ankle .
- the sole ordinate y foot can also be used to determine the first ordinate of the ankle portion, which is not limited in this embodiment.
- Step S1446-2 obtaining, according to the horizontal spatial distribution histogram, a valley value region near the first longitudinal coordinate of the crotch portion, and determining an ordinate corresponding to the valley region as a second ordinate of the crotch portion, wherein the The distance between the valley region and the first longitudinal coordinate of the crotch portion is less than a third preset distance threshold;
- Step S1446-3 According to the imaging characteristics of the contour image of the human crotch, the horizontal spatial distribution histogram near the position of the first ordinate of the ankle has a peak region, and the corresponding ordinate is y' chest . That is, the second ordinate of the ankle.
- Step S1446-4 obtaining the ordinate of the ankle according to the first ordinate of the ankle and the second ordinate of the ankle
- the expression is as follows:
- step S144 further includes:
- Step S1447 Acquire the knee longitudinal coordinate of the human body.
- Step S1447 specifically includes:
- Step S1447-1 obtaining the first vertical coordinate y of the human knee according to the top ordinate y head or the human sole ordinate y foot and the height height, the height and the knee ratio HEIGHT_KNEE_RATIO in the preset human body scale model. Knee .
- the expression of the first ordinate y knee of the ankle is as follows:
- y knee height*HEIGHT_KNEE_RATIO+y head ;
- HEIGHT_KNEE_RATIO is an average value obtained based on the measured values of the height of the human body and the horizontal position of the knee of the human body in the N-dimensional millimeter-wave grayscale image, and is preferably 0.78 in the present embodiment.
- the first ordinate of the knee is the approximate value of the knee ordinate.
- the sole ordinate y foot can also be used to determine the first ordinate of the knee, which is not limited in this embodiment.
- Step S1447-2 obtaining a peak region near the first longitudinal coordinate of the knee according to the horizontal spatial distribution histogram, and determining a vertical coordinate corresponding to the peak region as a second longitudinal coordinate of the knee, wherein the peak region and the The distance between the first ordinate of the knee is less than the fourth preset distance threshold;
- Step S1447-3 According to the imaging characteristics of the human knee contour image, the horizontal spatial distribution histogram near the position of the first ordinate of the knee has a peak region, and the corresponding ordinate is y' knee , that is The second ordinate of the ankle.
- Step S1447-4 obtaining the ordinate of the ankle according to the first ordinate of the ankle and the second ordinate of the ankle
- the expression is as follows:
- Step S1447-5 According to the knee longitudinal coordinate, the human body contour image below the ankle is divided to obtain a contour image of the human body.
- Embodiment 3 is a diagrammatic representation of Embodiment 3
- step S170 includes:
- Step S171 performing grayscale gradient feature extraction, edge feature extraction, average grayscale variance calculation, and edge smoothness calculation on the limb contour image.
- Step S172 Determine whether there is a foreign object in the limb contour image according to the grayscale gradient feature, the edge feature, the average grayscale variance, and the edge smoothness.
- step S172 includes:
- LBP feature threshold interval [LBP feature threshold minimum value LBP feature threshold maximum
- step S172 includes:
- step S180 includes:
- Step S181 identifying a location of the foreign object in the limb contour image
- Step S182 Generate a foreign matter detection result according to the position of the foreign object in the limb contour image.
- Embodiment 4 is a diagrammatic representation of Embodiment 4:
- the human body image mapping system provided in this embodiment includes:
- the human body grayscale image acquiring unit 110 is configured to acquire a millimeter wave grayscale image of the human body
- a human body contour image extracting unit 120 configured to extract a human body contour image in the millimeter wave grayscale image
- a histogram construction unit 130 configured to construct a vertical spatial distribution histogram of the human body contour image in a vertical direction and a horizontal spatial distribution histogram in a horizontal direction;
- a first limb position acquiring unit 140 configured to acquire a limb position of the human body in the human body contour image according to the vertical spatial distribution histogram, the horizontal spatial distribution histogram, and the preset human body scale model;
- a second limb position acquiring unit 150 configured to acquire a limb position of the human body in the cartoon image of the human body
- the mapping unit 160 is configured to construct a position mapping relationship between the human body contour image and each limb in the human body cartoon image according to the limb position of the human body in the human body contour image and the limb position of the human body in the human body cartoon image. .
- the human body grayscale image acquisition unit 110 may specifically include a millimeter wave data acquisition device (for example, a millimeter wave transceiver) and a millimeter wave imaging system (for example, a millimeter wave imager).
- a millimeter wave data acquisition device for example, a millimeter wave transceiver
- a millimeter wave imaging system for example, a millimeter wave imager
- the human body contour image extracting unit 120 is specifically configured to:
- the binarized image B(x, y) is subjected to a morphological operation in both the horizontal direction and the vertical direction to generate a human body contour image B1(x, y).
- the human body contour image extracting unit 120 is specifically configured to:
- the morphological expansion operation and the etching operation are performed on the B(x, y) image in the horizontal direction and the vertical direction.
- a foreign object recognition unit configured to identify a foreign object in the contour image of the human body according to a preset foreign object feature recognition model
- a first foreign object identification unit configured to identify a position of the foreign object on the human body contour image according to a limb position of the human body in the human body contour image
- a second foreign object identification unit configured to identify a position of the foreign object on the cartoon image of the human body according to the position mapping relationship, and generate a foreign object detection result.
- the foreign object recognition unit comprises:
- a metal foreign matter identification unit configured to determine a position of a metal foreign object according to an area in the human body contour image whose gray value is greater than a preset gray threshold and has a well-defined outline;
- a non-metallic foreign matter identification unit configured to determine a position of a non-metallic foreign object according to a geometrically complex and well-defined geometric region in the human body contour image
- the edge foreign matter identifying unit is configured to determine a position of the foreign object at the edge of the human body contour image according to a well-defined area of the millimeter wave gray image near the edge of the human body contour image.
- the embodiment of the invention extracts the human body contour image in the millimeter wave gray image by acquiring the millimeter wave gray image of the human body, and distributes the vertical space distribution histogram in the vertical direction and the horizontal spatial distribution in the horizontal direction according to the human body contour image.
- the histogram acquires the position of the human body, identifies the foreign object according to the preset foreign object feature recognition model, identifies the foreign object in the contour image of the human body, identifies the position of the foreign object on the contour image of the human body, and generates a foreign matter detection result, which can greatly improve the accuracy of the foreign matter detection.
- Sexual and identifiable metal foreign bodies and non-metallic foreign bodies are examples of the foreign matter detection.
- Embodiment 5 is a diagrammatic representation of Embodiment 5:
- this embodiment is a further refinement of the first limb position acquiring unit 140 in the fourth embodiment.
- the first limb position acquiring unit 140 includes:
- a vertical center line coordinate acquiring unit 141 configured to perform local maximum value detection on the vertical spatial distribution histogram, and acquire a vertical midline abscissa of the human body;
- a top, a head, and a sole coordinate acquiring unit 142 configured to perform local minimum detection on the horizontal spatial distribution histogram, and acquire a human hand's top ordinate, a top ordinate, and a sole ordinate;
- a height obtaining unit 143 configured to acquire a height of the human body according to the top ordinate and the sole ordinate;
- a limb contour image acquiring unit 144 configured to: according to any one of the top ordinate or the sole ordinate and the vertical midline abscissa, the height, the vertical spatial distribution histogram, and the horizontal spatial distribution
- the histogram and the preset human body scale model acquire image of the limb contour corresponding to each limb of the human body.
- the vertical center line coordinate acquiring unit 141 is specifically configured to:
- the upper half of the vertical distribution histogram is extracted as a sub-vertical distribution histogram.
- the top, top, and sole coordinate acquisition unit 142 specifically includes:
- a hand top coordinate acquiring unit configured to perform local minimum detection on the first region of the horizontal spatial distribution histogram, and determine a hand top ordinate according to the detected local minimum value
- a head coordinate acquiring unit configured to acquire a pixel number of a pixel value of each row of the second area of the horizontal spatial distribution histogram as a first preset pixel threshold, according to the number of the pixel points being smaller than the first preset a line of thresholds that determines the ordinate of the head of the human body;
- a sole coordinate acquiring unit configured to acquire a pixel number of a pixel value of each row of the third region of the horizontal spatial distribution histogram as a second preset pixel threshold, where the number of pixels is smaller than the second preset
- the threshold line determines the ordinate of the sole of the human body.
- the hand top coordinate acquisition unit is specifically used to:
- Performing local minimum detection on the first region of the horizontal spatial distribution histogram, and detecting the local minimum is the human hand top ordinate y hand ; wherein the first region is a closed interval [vertical midline abscissa x mid - first set value, vertical midline abscissa x mid + second set value].
- the overhead coordinate acquisition unit is specifically used to:
- the coordinates of the row of the line are determined as the y head of the human body, wherein the second region is a closed interval [vertical midline abscissa x mid - third set value, vertical midline abscissa x mid + fourth Set value].
- the sole coordinate acquisition unit is specifically used to:
- the coordinates of the row of the line are determined as the sole y foot of the human body, wherein the third region is a closed interval [between the bottom of the horizontal spatial distribution histogram and the bottom of the horizontal spatial distribution histogram The distance is the position of the fifth set value].
- the limb contour image acquisition unit 144 includes:
- a shoulder coordinate and head width acquisition unit for using any one of the overhead ordinate or the sole ordinate and the vertical midline abscissa, the height, the horizontal spatial distribution histogram, and a preset a human body scale model that obtains a shoulder ordinate of the human body and a width of a human body's head relative to the vertical midline abscissa;
- a head contour image acquiring unit configured to acquire an accurate width of the head according to the height, the body part proportional probability model, and the vertical spatial distribution histogram, and acquire a head contour image
- a left and right arm contour image acquiring unit configured to remove the head contour image according to image information above the shoulder in the human body contour image, and obtain a contour image of the left and right arms of the human body;
- a chest coordinate acquiring unit configured to obtain a first longitudinal coordinate of the human breast as the chest of the human body according to the top ordinate or the human sole ordinate and the height, the ratio of the height and the chest in the preset human body scale model The approximate value of the coordinates;
- a chest coordinate acquiring unit configured to obtain a first longitudinal coordinate of the human breast as the abdominal vertical of the human body according to the top ordinate or the human sole ordinate and the height, the ratio of the height and the abdomen in the preset human body scale model The approximate value of the coordinates;
- the ankle coordinate acquiring unit is configured to obtain the first longitudinal coordinate of the human body as the human body according to the top ordinate or the human sole ordinate and the height and the ratio of the height and the ankle in the preset human body scale model The approximate value of the ordinate of the ankle.
- a knee coordinate acquiring unit configured to obtain a first longitudinal axis of the human knee as a knee longitudinal of the human body according to the top ordinate or the human sole ordinate and the height, the ratio of the height and the knee in the preset human body scale model The approximate value of the coordinates.
- This embodiment is a further refinement of the foreign object identification unit and the foreign object identification unit on the basis of the fifth embodiment.
- the foreign object identification unit is specifically configured to:
- Gray contour feature extraction, edge feature extraction, average gray variance calculation, and edge smoothness calculation are performed on the limb contour image.
- Whether there is a foreign object in the limb contour image is determined according to the grayscale gradient feature, the edge feature, the average grayscale variance, and the edge smoothness.
- determining whether there is a foreign object in the contour image of the limb includes:
- LBP feature threshold interval [LBP feature threshold minimum value, LBP feature threshold maximum value]
- determining whether there is a foreign object in the limb contour image includes:
- the foreign object identification unit is specifically configured to:
- a foreign matter detection result is generated based on the position of the foreign matter in the limb contour image.
- a terminal device 200 includes: a processor 210, a memory 220, and a computer program stored in the memory 220 and executable on the processor 210, such as the above embodiment.
- Software method from one to three.
- the processor 210 implements the steps in the embodiments of the above-described respective human body image mapping methods when executing the computer program, such as steps S110 to S160 shown in FIG.
- the processor 210 implements the functions of the units in the various apparatus embodiments described above when executing the computer program, such as the functions of the units 110 to 160 shown in FIG.
- a computer program can be partitioned into one or more units, one or more units being stored in memory 220 and executed by processor 210 to perform the present invention.
- One or more of the units may be a series of computer program instructions that are capable of performing a particular function for describing the execution of the computer program in the terminal device 200.
- the computer program may be divided into an adult body grayscale image acquiring unit, a human body contour image extracting unit, a histogram building unit, a first limb position acquiring unit, a second limb position acquiring unit, and a mapping unit, and the specific functions of each module are as follows:
- a human body grayscale image acquiring unit for acquiring a millimeter wave grayscale image of the human body
- a human body contour image extracting unit configured to extract a human body contour image in the millimeter wave grayscale image
- a histogram construction unit configured to construct a vertical spatial distribution histogram of the human body contour image in a vertical direction and a horizontal spatial distribution histogram in a horizontal direction;
- a first limb position acquiring unit configured to acquire a limb position of the human body in the human body contour image according to the vertical spatial distribution histogram, the horizontal spatial distribution histogram, and a preset human body scale model;
- a second limb position acquiring unit configured to acquire a limb position of the human body in the cartoon image of the human body
- a mapping unit configured to construct a position mapping relationship between the human body contour image and each limb in the human body cartoon image according to a limb position of the human body in the human body contour image and a limb position of the human body in the human body cartoon image.
- the terminal device 200 can be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
- the terminal device 200 can include, but is not limited to, the processor 210, the memory 220. It will be understood by those skilled in the art that FIG. 12 is only an example of the terminal device 200, and does not constitute a limitation of the terminal device 200, and may include more or less components than those illustrated, or combine some components, or different components.
- the terminal device 200 may further include an input/output device, a network access device, a bus, and the like.
- the processor 210 may be a central processing unit (CPU), or may be another general-purpose processor 210, a digital signal processor (DSP), or an application specific integrated circuit (ASIC). ), a Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, and the like.
- the general purpose processor 210 can be the microprocessor 210 or the processor 210 can be any conventional processor 210 or the like.
- the memory 220 may be an internal storage unit of the terminal device 200, such as a hard disk or a memory of the terminal device 200.
- the memory 220 may also be an external storage device of the terminal device 200, such as a plug-in hard disk provided on the terminal device 200, a smart memory card (SMC), a Secure Digital (SD) card, and a flash memory card (Flash). Card) and so on.
- the memory 220 may also include both an internal storage unit of the terminal device 200 and an external storage device.
- the memory 22061 is used to store computer programs and other programs and data required by the terminal device 200.
- the memory 220 can also be used to temporarily store data that has been output or is about to be output.
- each functional unit and module in the foregoing system may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the integrated unit may be implemented by hardware.
- Formal implementation can also be implemented in the form of software functional units.
- the specific names of the respective functional units and modules are only for the purpose of facilitating mutual differentiation, and are not intended to limit the scope of protection of the present application.
- the disclosed device/terminal device 200 and method may be implemented in other manners.
- the device/terminal device 200 embodiment described above is merely illustrative.
- the division of a module or a unit is only a logical function division.
- there may be another division manner such as multiple units or Components can be combined or integrated into another system, or some features can be ignored or not executed.
- the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in electrical, mechanical or other form.
- the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
- each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
- the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
- Units if implemented as software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present invention implements all or part of the processes in the above embodiments, and can also be completed by a computer program to instruct related hardware.
- the computer program can be stored in a computer readable storage medium. When executed by the processor 210, the steps of the various method embodiments described above can be implemented. .
- the computer program includes computer program code, and the computer program code may be in the form of source code, object code form, executable file or some intermediate form.
- the computer readable medium may include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory 220, a Read-Only Memory (ROM) (ROM), and a random Access memory 220 (RAM, Random Access Memory), electrical carrier signals, telecommunications signals, software distribution media, and the like.
- ROM Read-Only Memory
- RAM Random Access Memory
- electrical carrier signals telecommunications signals
- software distribution media and the like.
- the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, computer readable media does not include It is an electrical carrier signal and a telecommunication signal.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
本方案涉及图像处理领域，提供一种人体图像映射方法、系统及终端设备，本发明通过获取人体的毫米波灰度图像，提取毫米波灰度图像中的人体轮廓图像，并根据人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图，获取人体轮廓图像中的人体的肢体位置，获取人体卡通图像中的人体的肢体位置，并构建人体轮廓图像和人体卡通图像中各肢体的位置映射关系，可以实现由人体轮廓图像到人体卡通图像的映射，保护用户隐私。
Description
本发明实施例属于图像处理技术领域，尤其涉及一种人体图像映射方法、系统及终端设备。
随着对暴恐分子的打击力度加大以及对人们日常出现的安全考虑，安检已成为人们在公共交通出行中必须进行的检查项目。目前基于毫米波的安检设备，由于对人体的辐射伤害较小，被广泛应用于案件领域，实现对人体的安全检查。
然而，现有的毫米波成像安检设备通常都是直接获取人体的全身图像，由于人体的全身图像中包括人体的隐私部位图像，可能会导致用户的隐私被泄露，造成一定的不良影响。
发明内容
本发明实施例提供一种人体图像映射方法、系统及终端设备，旨在解决目前的现有的毫米波安检设备通常都是直接获取人体的全身图像，由于人体的全身图像中包括人体的隐私部位图像，，可能会导致用户的隐私被泄露，造成一定的不良影响的问题。
本发明实施例第一方面提供一种人体图像映射方法，所述方法包括：
获取人体的毫米波灰度图像；
提取所述毫米波灰度图像中的人体轮廓图像；
构建所述人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图；
根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置；
获取人体卡通图像中的人体的肢体位置；
根据所述人体轮廓图像中的人体的肢体位置和所述人体卡通图像中的人体的肢体位置，构建所述人体轮廓图像和所述人体卡通图像中各肢体的位置映射关系。
本发明第二方面提供一种人体图像映射系统，所述系统包括：
人体灰度图像获取单元，用于获取人体的毫米波灰度图像；
人体轮廓图像提取单元，用于提取所述毫米波灰度图像中的人体轮廓图像；
直方图构建单元，用于构建所述人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图；
第一肢体位置获取单元，用于根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置；
第二肢体位置获取单元，用于获取人体卡通图像中的人体的肢体位置；
映射单元，用于根据所述人体轮廓图像中的人体的肢体位置和所述人体卡通图像中的人体的肢体位置，构建所述人体轮廓图像和所述人体卡通图像中各肢体的位置映射关系。
本发明实施例的第三方面提供一种终端设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述方法的步骤。
本发明实施例的第四方面提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现上述方法的步骤。
本发明实施例通过获取人体的毫米波灰度图像，提取毫米波灰度图像中的人体轮廓图像，并根据人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布 直方图，获取人体轮廓图像中的人体的肢体位置，获取人体卡通图像中的人体的肢体位置，并构建人体轮廓图像和人体卡通图像中各肢体的位置映射关系，可以实现由人体轮廓图像到人体卡通图像的映射，保护用户隐私。
图1是本发明实施例一提供的人体图像映射方法的基本流程框图；
图2是本发明实施例一提供的毫米波灰度图像；
图3是本发明实施例一提供的二值化图像；
图4是本发明实施例一提供的人体轮廓图像；
图5是本发明实施例一提供的垂直空间分布直方图；
图6是本发明实施例一提供的水平空间分布直方图；
图7是本发明实施例一提供的人体轮廓图像和人体卡通图像的映射关系示意图；
图8是本发明实施例一提供的异物检测结果示意图；
图9是本发明实施例二提供的人体图像映射方法的基本流程框图；
图10是本发明实施例四提供的人体图像映射系统的结构框图；
图11是本发明实施例五提供的第一肢体位置获取单元的结构示意图；
图12是本发明实施例七提供的终端设备的结构示意图。
为了使本技术领域的人员更好地理解本发明方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚地描述，显然，所描述的实施例是本发明一部分的实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语“包括”以及它们任何变形，意图在于覆盖不排他的包含。例如包含一系列步骤或单元的过程、方法或系统、产品或设备没有限定于已列出的步骤、单元或单元，而是可选地还包括没有列出的步骤、单元或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤、单元或单元。
实施例一：
如图1所示，本实施例提供一种人体图像映射方法，其包括：
步骤S110：获取人体的毫米波灰度图像。
在具体应用中，可以使人体双手上扬举过头顶或双手平举至与肩部同高的位置站立，也可以使人体双手自然下垂站立，或者采用其他符合安检标准的站姿均可，本发明实施例不对人体的站姿作特别限定，然后利用毫米波数据采集设备(例如，毫米波收发机)获取人体正面或背面的毫米波数据，并利用毫米波成像系统(例如，毫米波成像仪)将人体的毫米波数据处理为人体正面或背面的毫米波灰度图像。
如图2所示为成年男性双手上扬高举过头顶拍摄得到的人体的毫米波灰度图像。
步骤S120：提取所述毫米波灰度图像中的人体轮廓图像。
在具体应用中，所述人体轮廓图像，是指与人体的肢体轮廓形状对应的图像。
在一个实施例中，步骤S120具体包括：
步骤S121：对毫米波灰度图像I(x，y)进行灰度分割，获取毫米波灰度图像I(x，y)的二值化图像B(x，y)。图3示例性的示出二值化图像B(x，y)的示意图。
在具体应用中，步骤S121中对毫米波图像灰度I(x，y)进行灰度分割，获得对应二值化图像B(x，y)可以根据下式实现：
其中，x为图像列数目，y为图像行数目，X为图像最大列数目，Y为图像最大行数目，T为阈值，B(x，y)中数值为255的代表人体区域。
步骤S122：对二值化图像B(x，y)在水平方向上和垂直方向上均进行形态学操作，生成人体轮廓图像B1(x，y)，该人体轮廓图像B1(x，y)如图4所示。
步骤S122中的水平方向和垂直方向分别是指，与人体身高方向垂直和平行的方向。
在具体应用中，步骤S122具体包括：
对B(x，y)图像在水平方向上和垂直方向上进行形态学膨胀操作和腐蚀操作，所述膨胀操作的核函数大小为1×3，所述腐蚀操作的核函数大小为3×1大小，所述膨胀操作和所述腐蚀操作的表达式分别为：
B1(x，y)＝max
(x′,y′)：element(x′,y′)≠0B(x+x′,y+y′)(膨胀操作)；
B1(x，y)＝min
(x′,y′)：element(x′,y′)≠0B(x+x′,y+y′)(腐蚀操作)；
其中，x′和y′表示核函数所对应的平移单位数值。
在具体应用中，在进行上述形态学操作时，可以对二值化图像B(x，y)先进行膨胀操作后进行腐蚀操作，以获得人体轮廓图像B1(x，y)；也可以对二值化图像B(x，y)先进行腐蚀操作后进行膨胀操作，获得人体轮廓图像B1(x，y)。为了得到边缘平滑的人体轮廓图像，在进行形态学操作时，本发明优选为对二值化图像B(x，y)先进行膨胀操作后进行腐蚀操作，获得边缘平滑的人体轮廓图像B1(x，y)。
步骤S130：构建所述人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图。
本实施例中的空间分布直方图，是指按空间位置统计人体轮廓图像的灰度值信息所得到的直方图。垂直空间分布直方图具体是指：以从左到右(从左到右即为，当人体站立水平面上，从人体左侧到右侧且平行于水平面的方向)的空间位置为横坐标，人体轮廓图像的灰度值在单位空间位置长度内出现的频率为纵坐标的直方图。同理，水平空间分布直方图具体是指：以从上到下(从上到小即为，当人体站立在水平面上，从人体头顶到脚底且垂直于水平面的方向)的空间位置为纵坐标，人体轮廓图像的灰度值在单位空间位置长度内出现的频率为横坐标的直方图。
在具体应用中，所述垂直空间分布直方图H如图5所示，步骤S130中，构建人体轮廓图像B1(x，y)在垂直方向上的垂直空间分布直方图H，具体可以通过如下公式实现：
对垂直空间分布直方图H进行平滑处理，平滑尺度设定为3的公式如下:
在具体应用中，所述水平空间分布直方图V如图6所示，步骤S130中，构建人体轮廓图像B1(x，y)在水平方向上的水平空间分布直方图V，具体可以通过如下公式实现：
对垂直空间分布直方图V进行平滑处理，平滑尺度设定为3的公式如下：
步骤S140：根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置。
在具体应用中，所述人体比例模型根据人体的肢体结构、肢体之间的大小比例和肢体形状等特点构造。
在具体应用中，所述肢体位置具体包括人体的头部、肩部、胸部、腹部、裆部、四肢(双手、双脚)和膝关节位置，在特殊情况下，对于肢体残疾人士，肢体位置对应缺失某一部分数据。
步骤S150：获取人体卡通图像中的人体的肢体位置。
在具体应用中，人体卡通图像可以是事先通过计算机模拟出来的仿人形图像，该图像中的各肢体的位置坐标均是可以事先设定的，也就是说人体卡通图像中的人体的肢体位置均是可以由工作人员事先设定的已知数据，在需要使用该已知数据时，只需要直接调用事先设定并存储的人体卡通图像中的人体的肢体位置的数据即可。在其他应用中，人体卡通图像中的人体的肢体位置也可以通过与获取人体轮廓图像中的人体的肢体位置相同的方式获得，具体的，可以通过下列步骤来实现：
提取人体卡通图像中的人体轮廓图像；
构建所述人体卡通图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图；
根据预设人体比例模型以及所述人体卡通图像的垂直空间分布直方图和水平空间分布直方图，获取所述人体卡通图像中的人体的肢体位置。
步骤S160：根据所述人体轮廓图像中的人体的肢体位置和所述人体卡通图像中的人体的肢体位置，构建所述人体轮廓图像和所述人体卡通图像中各肢体的位置映射关系。
在本实施例中，各肢体的位置映射关系即是指各肢体的位置坐标的映射关系。例如，假设人体轮廓图像中的头部位置坐标为(A，B)，人体卡通图像中的头部位置坐标为(a，b)，则头部的位置映射关系即为坐标(A，B)到(a，b)的映射。应当理解的是，上述举例中的头部位置坐标仅仅只是示例性的，由于在实际应用中，每个肢体在图像中都是以二维平面图像的形式显示的，因此每个肢体对应的位置坐标可以不止一个，在某些情况下可以将肢体图像的几何中心的位置坐标作为该肢体的位置坐标，也可以在肢体图上选取二个或以上的坐标点，将选取的坐标点的位置坐标作为该肢体的位置坐标。
在一个实施例中，步骤S160之前包括：
将所述人体轮廓图像划分为预设数量的第一图像区域；
将所述人体卡通图像划分为预设数量的第二图像区域；
建立所述第一图像区域和所述第二图像区域之间的位置映射关系。
在具体应用中，在建立两个图像中各肢体的精确位置坐标的映射关系之前，可以通过相同的图像区域划分方式，将两个图像划分为数量相等的图像区域，通过建立两个图像中的图像区域之间的映射关系，可以粗略的实现两个图像中肢体位置的对应。在一个实施例中，若要进一步的实现肢体位置的精确对应，则可以在图像区域的映射关系的基础上，通过步骤S160进一步的将各肢体的位置坐标对应起来。在其他实施例中，还可以通过进一步对两个图像进行精细划分，来实现更为精确的位置对应，具体的可以通过增加第一图像区域和第二图像区域的数量来实现。
如图7所示，示例性的示出了将人体轮廓图像和人体卡通图像均划分为9个图像区域时的示意图，图中标号相同的图像区域具有映射关系。图7中还示例性的示出了两个位置坐标相对应的坐标点，分别表示为点A和点a，图中示出了两个坐标点与其各自所在图像区域5的边界之间的距离，以及图像区域5在两个图像中的边长分别为X0和Y0、X1和Y1，假设A的坐标为(x0，y0)，a的坐标为(x1，y1)，则由于X0和Y0、X1和Y1以及A的坐标为(x0，y0)均是已知的，可以求得x1＝X1*x0/X0，y1＝Y1*x0/Y0，从而得到a点相对于A点的坐标，即可得到两个坐标点之间的映射关系；同理，可以求得人体轮廓图像和人体卡通图像中任意两点之间的映射关系。
在一个实施例，所述人体图像映射方法还包括：
步骤S170：根据预设异物特征识别模型，识别所述人体轮廓图像中的异物。
在具体应用中，所述异物可以是金属类的枪支、刀具、金块等金属类异物，也可以是化学药剂、象牙、玉石等非金属类异物。
在一实施例中，步骤S170包括：
根据所述人体轮廓图像中灰度值大于预设灰度阈值且轮廓分明的区域，确定金属类异物的位置；
根据所述人体轮廓图像中纹理复杂、轮廓分明的几何图形区域，确定非金属类异物的位置；
根据所述毫米波灰度图像中，靠近所述人体轮廓图像边缘的轮廓分明的区域，确定所述人体轮廓图像边缘的异物的位置。
通过上述异物识别方法可大大提高异物检测的准确性且可识别金属类异物和非金属类异物。
步骤S180：根据所述人体轮廓图像中的人体的肢体位置，标识所述异物在所述人体轮廓图像上的位置。
在具体应用中，标识所述异物在所述人体轮廓图像上的位置，具体是指在各肢体所对应的人体轮廓图像位置上，标识出异物，例如当异物在人体腹部，则在人体轮廓图像的腹部所在位置标识出该异物。
步骤S190：根据所述位置映射关系，标识所述异物在所述人体卡通图像上的位置并生成异物检测结果。
在具体应用中，步骤S190和步骤S180的标识异物的方法相同，所不同的仅仅只是将人体轮廓图像替换为人体卡通图像。
如图8所示，即为检测到异物在人体腹部和左腿的大腿正前方时，直接在人体轮廓图像中的腹部位置和大腿正前方的位置处用明显的矩形框标识出异物的位置。
在具体应用中，所述异物检测结果可以以文字、图像、语音、表格等形式呈现，例如，当人体胸部有异物，可用文字呈现“胸部”或更具体的“胸部有金属类异物”。
本实施例通过获取人体的毫米波灰度图像，提取毫米波灰度图像中的人体轮廓图像，并根据人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图，获取人体轮廓图像中的人体的肢体位置，获取人体卡通图像中的人体的肢体位置，并构建人体轮廓图像和人体卡通图像中各肢体的位置映射关系，可以实现由人体轮廓图像到人体卡通图像的映射，保护用户隐私。
实施例二：
如图9所示，本实施例是对实施例一中步骤S140的进一步细化，在本实施例中，步骤S140包括：
步骤S141：对所述垂直空间分布直方图进行局部最大值检测，获取人体的垂直中线横坐标。
在一个实施例中，步骤S141具体包括：
步骤S1411：提取垂直分布直方图H的上半部分，作为子垂直分布直方图H
1。
在具体应用中，子垂直分布直方图H
1为垂直分布直方图H的上半部分，即按照50％*H的方式提取，本实施例还可以按照其他比例提取垂直分布直方图H的上半部分，本实施例中并不对该提取比例作限定。
步骤S1412：获取子垂直分布直方图H
1的最大值。该最大值的获取公式如下：
其中，x
mid所在列即为所述垂直中线横坐标所在的位置，如图4所示中所示的穿过人体头部、胸部、腹部、裆部的垂直于多条水平线的那条线即为偶数垂直中线。
在具体应用中，根据所述垂直分布直方图获得所述垂直中线横坐标也可以采用其他方法 实现，在此不予示例。
步骤S142：对所述水平空间分布直方图进行局部最小值检测，获取人体的手顶纵坐标、头顶纵坐标和脚底纵坐标。
在一个实施例中，步骤S142具体包括：
步骤S1421：对所述水平空间分布直方图的第一区域进行局部最小值检测，根据检测到的局部最小值所在行，确定人体的手顶纵坐标。
在具体应用中，步骤S1421具体为：
对所述水平空间分布直方图的第一区域进行局部最小值检测，检测到的局部最小值即为人体的手顶纵坐标y
hand；其中，所述第一区域为闭区间[垂直中线横坐标x
mid-第一设定值，垂直中线横坐标x
mid+第二设定值]。
在具体应用中，第一设定值和第二设定值可以相同、也可以不相同且可根据用户的实际需要设定。例如，将所述第一区域的大小定义为101像素，第一设定值和第二设定值相同，则第一局部区域为[x
mid-50，x
mid+50]。
步骤S1422：获取所述水平空间分布直方图的第二区域的每一行中像素值为第一预设像素阈值的像素点个数，根据所述像素点个数小于第一预设阈值的行，确定人体的头顶纵坐标。
在具体应用中，步骤S1422具体为：
在所述水平空间分布直方图的第二区域内，获取所述第二区域内每一行中像素值为第一预设像素阈值的像素点个数，将像素点个数小于第一预设阈值的那一行的所在的坐标确定为人体的头顶纵坐标y
head，其中，所述第二区域为闭区间[垂直中线横坐标x
mid-第三设定值，垂直中线横坐标x
mid+第四设定值]。
在具体应用中，第三设定值和第四设定值可以相同、也可以不相同且可根据用户的实际需要设定。例如，将所述第二区域的大小定义为31像素，第三设定值和第四设定值相同，则第二局部区域为[x
mid-15，x
mid+15]。
步骤S1423：获取所述水平空间分布直方图的第三区域的每一行中像素值为第二预设像素阈值的像素点个数，根据所述像素点个数小于第二预设阈值的行，确定人体的脚底纵坐标。
在具体应用中，步骤S1423具体为：
在所述水平空间分布直方图的第三区域内，获取所述第三区域内每一行中像素值为第二预设像素阈值的像素点个数，将像素点个数小于第二预设阈值的那一行的所在的坐标确定为人体的脚底纵坐标y
foot，其中，所述第三区域为闭区间[所述水平空间分布直方图的底部，与所述水平空间分布直方图的底部之间的距离为第五设定值的位置]。
在具体应用中，所述第五设定值可以根据实际需要设置。例如，将所述第三区域的大小定义为60像素，则所述水平空间分布直方图底部到距离该底部60像素的位置之间的区域。在具体应中，所述第二预设像素阈值可以设置为255，在获得每一行中像素值小于255的像素点个数时，可以在所述第三区域内从上到下检测每一行图像的中像素值为255的像素点，并统计每一行中像素值为255像素点个数，所述像素点个数小于预设阈值的行，即为所述脚底纵坐标y
foot所在的位置。
步骤S143：根据所述头顶纵坐标和所述脚底纵坐标，获取人体的身高。
在一个实施例中，步骤S143中获取人体的身高height的公式为：
height＝|y
foot-y
head|。
步骤S144：根据所述头顶纵坐标或所述脚底纵坐标以及所述垂直中线横坐标、所述身高、所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取与人体的各肢体对应的肢体轮廓图像。
在一个实施例中，步骤S144包括：
步骤S1441：根据所述头顶纵坐标或所述脚底纵坐标中的任一个以及所述垂直中线横坐标、所述身高、所述水平空间分布直方图和预设人体比例模型，获得人体的肩部纵坐标和相对于所述垂直中线横坐标的人体的头部的宽度；
步骤S1442：根据所述身高、所述人体部位比例概率模型和所述垂直空间分布直方图，获取所述头部的精确宽度，并获取头部轮廓图像；
步骤S1443：根据所述人体轮廓图像中肩部以上的图像信息，去除所述头部轮廓图像，获得人体的左右双臂轮廓图像。
在具体应用中，可对所述左右双臂轮廓图像进行边缘检测操作，以获得人体的左右双臂的边缘轮廓，根据人体的双臂的形状特点和轮廓的曲率变化规律，可以获得左右双臂的手肘部位的坐标，对所述左右手肘部位的拐点位置进行切割，可以分别获得左右双臂的大臂轮廓图像和小臂轮廓图像。
在具体应用中，可根据所述头顶纵坐标或所述脚底纵坐标以及所述身高、所述水平空间分布直方图和预设人体比例模型，获得人体的肩部坐标。
步骤S1441具体包括：
步骤S1441-1：根据所述头顶纵坐标y
head或所述脚底纵坐标y
foot以及所述身高height、所述预设人体比例模型中的人体的身高和胸部的比值HEIGHT_CHEST_RATIO，获得人体的肩部第一纵坐标y
shoulder。当采用所述头顶纵坐标y
head计算时，所述肩部第一纵坐标y
shoulder的表达式如下：
y
shoulder＝height*HEIGHT_CHEST_RATIO+y
head；
其中，HEIGHT_CHEST_RATIO是基于N幅人体毫米波灰度图像中人体的身高和人体的肩部水平位置的测量值得到的平均值，本实施例中，优选HEIGHT_CHEST_RATIO为0.2。所述肩部第一纵坐标y
shoulder即为所求的所述肩部纵坐标的接近值。
在具体应用中，还可以采用所述脚底纵坐标y
foot计算，以确定人体的肩部第一纵坐标，本实施例不对此进行限定。
步骤S1441-2：根据所述水平空间分布直方图，获得所述肩部第一纵坐标附近的峰值区域，将峰值区域对应的纵坐标确定为肩部第二纵坐标，其中，所述峰值区域与所述胸部第一纵坐标的距离小于第一预设距离阈值。
步骤S1441-3：根据人体肩部的成像特点，在肩部纵坐标位置附近的水平空间分布直方图会有一个峰值区域，其所对应的纵坐标为y′
shoulder，即人体肩部第二纵坐标。
步骤S1441-4：根据所述肩部第一纵坐标和所述肩部第二纵坐标获得所述肩部纵坐标
的表达式如下：
步骤S1443包括：
S1443-1：根据人体左右手臂的灰度图像呈现不连续斑块状且人体组织成像面积较大的特点，获取成像面积较大的斑块区域的作为双臂轮廓图像所在区域；
S1443-2：计算所述斑块区域中第一成像面积和第二成像面积相对于所述斑块区域面积的比值，若所述比值在预设区间内，则所述第一成像面积和所述第二成像面积作为所述双臂轮廓图像，若所述比值不在预设区间内，则表示所述第一成像面积包含所述双臂轮廓图像，根据所述第一成像面积拟合的外接多边形的曲率变化进行切分，获得所述大臂轮廓图像和所述小臂轮廓图像。
在具体应用中，所述预设区间优选为[1，10]。
在一个实施例中，步骤S144还包括：
S1444：获取人体的胸部纵坐标。
步骤S1444包括：
步骤S1444-1：根据所述头顶纵坐标y
head或人体脚底纵坐标y
foot以及所述身高height、所述预设人体比例模型中的身高和胸部的比值HEIGHT_CHEST_RATIO，获得人体胸部第一 纵坐标y
chest。当采用所述头顶纵坐标y
head计算时，所述胸部第一纵坐标的表达式如下：
y
chest＝height*HEIGHT_CHEST_RATIO+y
head；
其中，HEIGHT_CHEST_RATIO是基于N幅人体的毫米波灰度图像中人体的身高和人体的胸部水平位置的测量值得到的平均值，本实施例中优选为0.8。胸部第一纵坐标y
chest即为所述胸部纵坐标的接近值。
具体应用中，还可以采用所述脚底纵坐标y
foot计算，以确定所述胸部第一纵坐标，本实施例不对此进行限定。
步骤S1444-2：根据所述水平空间分布直方图，获得所述胸部第一纵坐标附近的峰值区域，将峰值区域对应的纵坐标确定为胸部第二纵坐标，其中，所述峰值区域与所述胸部第一纵坐标的距离小于第二预设距离阈值；
步骤S1444-3：根据人体胸部轮廓图像的成像特点，在所述胸部第一纵坐标所在位置附近的水平空间分布直方图会有一个峰值区域，其所对应的纵坐标为y′
chest，即为所述胸部第二纵坐标。
步骤S1444-4：根据所述胸部第一纵坐标和所述胸部第二纵坐标获得所述胸部纵坐标
表达式如下：
在一个实施例中，步骤S144还包括：
步骤S1445：获取人体的腹部纵坐标。
步骤S1445具体包括：
根据所述头顶纵坐标y
head或人体脚底纵坐标y
foot以及所述身高height、所述预设人体比例模型中的身高和腹部的比值HEIGHT_ABDOMEN_RATIO，获得人体胸部第一纵坐标y
abdomen。当采用所述头顶纵坐标y
head计算时，所述腹部第一纵坐标的表达式如下：
y
abdomen＝height*HEIGHT_ABDOMEN_RATIO+y
head；
其中，HEIGHT_ABDOMEN_RATIO是基于N幅人体毫米波灰度图像中人体的身高和人体的腹部水平位置的测量值得到的平均值，本示例中优选为0.44。人体腹部第一纵坐标y
abdomen即为得到的所述腹部纵坐标的接近值。
体应用中，还可以采用所述脚底纵坐标y
foot计算，以确定所述腹部第一纵坐标，本实施例不对此进行限定。
在一个实施例中，步骤S144还包括：
步骤S1446：获取人体的裆部纵坐标。
步骤S1446具体包括：
步骤S1446-1：根据所述头顶纵坐标y
head或人体脚底纵坐标y
foot以及所述身高height、所述预设人体比例模型中的身高和裆部的比值HEIGHT_CROTCH_RATIO，获得人体裆部第一纵坐标y
crotch。当采用所述头顶纵坐标y
head计算时，所述裆部第一纵坐标y
crotch的表达式如下：
y
crotch＝height*HEIGHT_CROTCH_RATIO+y
head；
其中，HEIGHT_CROTCH_RATIO是基于N幅人体毫米波灰度图像中人体的身高和人体的裆部水平位置的测量值得到的平均值，本实施例中优选为0.51。所述裆部第一纵坐标y
crotch即为所述裆部纵坐标的接近值。
具体应用中，还可以采用所述脚底纵坐标y
foot计算，以确定所述裆部第一纵坐标，本实施例不对此进行限定。
步骤S1446-2：根据所述水平空间分布直方图，获得所述裆部第一纵坐标附近的谷值区域，将谷值区域对应的纵坐标确定为裆部第二纵坐标，其中，所述谷值区域与所述裆部第一纵坐标的距离小于第三预设距离阈值；
步骤S1446-3：根据人体裆部轮廓图像的成像特点，在所述裆部第一纵坐标所在位置附近的水平空间分布直方图会有一个峰值区域，其所对应的纵坐标为y′
chest，即为所述裆部第二纵坐标。
步骤S1446-4：根据所述裆部第一纵坐标和所述裆部第二纵坐标获得所述裆部纵坐标
的表达式如下：
在一实施例中，步骤S144还包括：
步骤S1447：获取人体的膝盖纵坐标。
步骤S1447具体包括：
步骤S1447-1：根据所述头顶纵坐标y
head或人体脚底纵坐标y
foot以及所述身高height、所述预设人体比例模型中的身高和膝盖的比值HEIGHT_KNEE_RATIO，获得人体膝盖第一纵坐标y
knee。当采用所述头顶纵坐标y
head计算时，所述裆部第一纵坐标y
knee的表达式如下：
y
knee＝height*HEIGHT_KNEE_RATIO+y
head；
其中，HEIGHT_KNEE_RATIO是基于N幅人体毫米波灰度图像中人体的身高和人体的膝盖水平位置的测量值得到的平均值，本实施中优选为0.78。所述膝盖第一纵坐标即为所述膝盖纵坐标的接近值。
具体应用中，还可以采用所述脚底纵坐标y
foot计算，以确定所述膝盖第一纵坐标，本实施例不对此进行限定。
步骤S1447-2：根据所述水平空间分布直方图，获得所述膝盖第一纵坐标附近的峰值区域，将峰值区域对应的纵坐标确定为膝盖第二纵坐标，其中，所述峰值区域与所述膝盖第一纵坐标的距离小于第四预设距离阈值；
步骤S1447-3：根据人体膝盖轮廓图像的成像特点，在所述膝盖第一纵坐标所在位置附近的水平空间分布直方图会有一个峰值区域，其所对应的纵坐标为y′
knee，即为所述裆部第二纵坐标。
步骤S1447-4：根据所述裆部第一纵坐标和所述裆部第二纵坐标获得所述裆部纵坐标
的表达式如下：
步骤S1447-5：根据所述膝盖纵坐标，对裆部以下的人体轮廓图像进行分割，获得人体的双腿轮廓图像。
实施例三：
本实施例在实施例二的基础上是对实施例一中步骤S170和S180的进一步细化，在本实施例中，步骤S170包括：
步骤S171：对所述肢体轮廓图像进行灰度梯度特征提取、边缘特征提取、平均灰度方差计算和边缘平滑度计算。
步骤S172：根据所述灰度梯度特征、所述边缘特征、所述平均灰度方差和所述边缘平滑度，判断所述肢体轮廓图像中是否有异物。
在一个实施例中，步骤S172包括：
判断所述肢体轮廓图像的当前LBP特征值是否在预设LBP特征阈值区间内，所述LBP 特征阈值区间为[LBP特征阈值最小值
LBP特征阈值最大值
若是，则判定所述肢体轮廓图像中无异物，否则，判定所述肢体轮廓图像中有异物。
根据上述方法，依次判断各肢体所对应的肢体轮廓图像中是否有异物。
在另一实施例中，步骤S172包括：
判断所述肢体轮廓图像的当前轮廓曲率特征值是否在预设轮廓曲率特征阈值区间内，所述轮廓曲率特征阈值区间为[轮廓曲率特征阈值最小值
轮廓曲率特征阈值最大值
若是，则判定所述肢体轮廓图像中无异物，否则，判定所述肢体轮廓图像中有异物。
在本实施例中，步骤S180包括：
步骤S181：在所述肢体轮廓图像中标识出所述异物的位置；
步骤S182：根据所述异物在所述肢体轮廓图像中的位置，生成异物检测结果。
实施例四：
如图10所示，本实施例提供的人体图像映射系统，其包括：
人体灰度图像获取单元110，用于获取人体的毫米波灰度图像；
人体轮廓图像提取单元120，用于提取所述毫米波灰度图像中的人体轮廓图像；
直方图构建单元130，用于构建所述人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图；
第一肢体位置获取单元140，用于根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置；
第二肢体位置获取单元150，用于获取人体卡通图像中的人体的肢体位置；
映射单元160，用于根据所述人体轮廓图像中的人体的肢体位置和所述人体卡通图像中的人体的肢体位置，构建所述人体轮廓图像和所述人体卡通图像中各肢体的位置映射关系。
在具体应用中，人体灰度图像获取单元110具体可以包括毫米波数据采集设备(例如，毫米波收发机)和毫米波成像系统(例如，毫米波成像仪)。
在一个实施例中，人体轮廓图像提取单元120具体用于：
对毫米波灰度图像I(x，y)进行灰度分割，获取毫米波灰度图像I(x，y)的二值化图像B(x，y)；
对二值化图像B(x，y)在水平方向上和垂直方向上均进行形态学操作，生成人体轮廓图像B1(x，y)。
在具体应用中，人体轮廓图像提取单元120具体还用于：
对B(x，y)图像在水平方向上和垂直方向上进行形态学膨胀操作和腐蚀操作。
在一个实施例中所述系统还包括：
异物识别单元，用于根据预设异物特征识别模型，识别所述人体轮廓图像中的异物；
第一异物标识单元，用于根据所述人体轮廓图像中的人体的肢体位置，标识所述异物在所述人体轮廓图像上的位置；
第二异物标识单元，用于根据所述位置映射关系，标识所述异物在所述人体卡通图像上的位置并生成异物检测结果。
在一实施例中，异物识别单元包括：
金属类异物识别单元，用于根据所述人体轮廓图像中灰度值大于预设灰度阈值且轮廓分明的区域，确定金属类异物的位置；
非金属类异物识别单元，用于根据所述人体轮廓图像中纹理复杂、轮廓分明的几何图形区域，确定非金属类异物的位置；
边缘异物识别单元，用于根据所述毫米波灰度图像中，靠近所述人体轮廓图像边缘的轮廓分明的区域，确定所述人体轮廓图像边缘的异物的位置。
本发明实施例通过获取人体的毫米波灰度图像，提取毫米波灰度图像中的人体轮廓图像，并根据人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布 直方图，获取人体的肢体位置，根据预设异物特征识别模型，识别所述人体轮廓图像中的异物，标识出异物在人体轮廓图像上的位置并生成异物检测结果，可大大提高异物检测的准确性且可识别金属类异物和非金属类异物。
实施例五：
如图11所示，本实施例是对实施例四中第一肢体位置获取单元140的进一步细化，该第一肢体位置获取单元140包括：
垂直中线坐标获取单元141，用于对所述垂直空间分布直方图进行局部最大值检测，获取人体的垂直中线横坐标；
手顶、头顶、脚底坐标获取单元142，用于对所述水平空间分布直方图进行局部最小值检测，获取人体的手顶纵坐标、头顶纵坐标和脚底纵坐标；
身高获取单元143，用于根据所述头顶纵坐标和所述脚底纵坐标，获取人体的身高；
肢体轮廓图像获取单元144，用于根据所述头顶纵坐标或所述脚底纵坐标中的任一个以及所述垂直中线横坐标、所述身高、所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取与人体的各肢体对应的肢体轮廓图像。
在一个实施例中，垂直中线坐标获取单元141具体用于：
提取所述垂直分布直方图的上半部分，作为子垂直分布直方图。
获取所述子垂直分布直方图的最大值。
在一个实施例中，手顶、头顶、脚底坐标获取单元142具体包括：
手顶坐标获取单元，用于对所述水平空间分布直方图的第一区域进行局部最小值检测，根据检测到的局部最小值所在的行，确定人体的手顶纵坐标；
头顶坐标获取单元，用于获取所述水平空间分布直方图的第二区域的每一行中像素值为第一预设像素阈值的像素点个数，根据所述像素点个数小于第一预设阈值的行，确定人体的头顶纵坐标；
脚底坐标获取单元，用于获取所述水平空间分布直方图的第三区域的每一行中像素值为第二预设像素阈值的像素点个数，根据所述像素点个数小于第二预设阈值的行，确定人体的脚底纵坐标。
在具体应用中，手顶坐标获取单元，具体用于：
对所述水平空间分布直方图的第一区域进行局部最小值检测，检测到的局部最小值即为人体的手顶纵坐标y
hand；其中，所述第一区域为闭区间[垂直中线横坐标x
mid-第一设定值，垂直中线横坐标x
mid+第二设定值]。
在具体应用中，头顶坐标获取单元，具体用于：
在所述水平空间分布直方图的第二区域内，获取所述第二区域内每一行中像素值为第一预设像素阈值的像素点个数，将像素点个数小于第一预设阈值的那一行的所在的坐标确定为人体的头顶纵坐标y
head，其中，所述第二区域为闭区间[垂直中线横坐标x
mid-第三设定值，垂直中线横坐标x
mid+第四设定值]。
在具体应用中，脚底坐标获取单元，具体用于：
在所述水平空间分布直方图的第三区域内，获取所述第三区域内每一行中像素值为第二预设像素阈值的像素点个数，将像素点个数小于第二预设阈值的那一行的所在的坐标确定为人体的脚底纵坐标y
foot，其中，所述第三区域为闭区间[所述水平空间分布直方图的底部，与所述水平空间分布直方图的底部之间的距离为第五设定值的位置]。
在一个实施例中，肢体轮廓图像获取单元144包括：
肩部坐标和头部宽度获取单元，用于根据所述头顶纵坐标或所述脚底纵坐标中的任一个以及所述垂直中线横坐标、所述身高、所述水平空间分布直方图和预设人体比例模型，获得人体的肩部纵坐标和相对于所述垂直中线横坐标的人体的头部的宽度；
头部轮廓图像获取单元，用于根据所述身高、所述人体部位比例概率模型和所述垂直空间分布直方图，获取所述头部的精确宽度，并获取头部轮廓图像；
左右双臂轮廓图像获取单元，用于根据所述人体轮廓图像中肩部以上的图像信息，去除所述头部轮廓图像，获得人体的左右双臂轮廓图像；
胸部坐标获取单元，用于根据所述头顶纵坐标或人体脚底纵坐标以及所述身高、所述预设人体比例模型中的身高和胸部的比值，获得人体胸部第一纵坐标作为人体的胸部纵坐标的接近值；
胸部坐标获取单元，用于根据所述头顶纵坐标或人体脚底纵坐标以及所述身高、所述预设人体比例模型中的身高和腹部的比值，获得人体胸部第一纵坐标作为人体的腹部纵坐标的接近值；
裆部坐标获取单元，用于根据所述头顶纵坐标或人体脚底纵坐标以及所述身高、所述预设人体比例模型中的身高和裆部的比值，获得人体裆部第一纵坐标作为人体的裆部纵坐标的接近值。
膝盖坐标获取单元，用于根据所述头顶纵坐标或人体脚底纵坐标以及所述身高、所述预设人体比例模型中的身高和膝盖的比值，获得人体膝盖第一纵坐标作为人体的膝盖纵坐标的接近值。
实施例六：
本实施例是在实施例五的基础上对异物标识单元和异物识别单元的进一步细化。
在本实施例中，异物标识单元具体用于：
对所述肢体轮廓图像进行灰度梯度特征提取、边缘特征提取、平均灰度方差计算和边缘平滑度计算。
根据所述灰度梯度特征、所述边缘特征、所述平均灰度方差和所述边缘平滑度，判断所述肢体轮廓图像中是否有异物。
在一个实施例中，判断所述肢体轮廓图像中是否有异物，具体包括：
判断所述肢体轮廓图像的当前LBP特征值是否在预设LBP特征阈值区间内，所述LBP特征阈值区间为[LBP特征阈值最小值，LBP特征阈值最大值]；
若是，则判定所述肢体轮廓图像中无异物，否则，判定所述肢体轮廓图像中有异物。
根据上述方法，依次判断各肢体所对应的肢体轮廓图像中是否有异物。
在另一实施例中，判断所述肢体轮廓图像中是否有异物，具体包括：
判断所述肢体轮廓图像的当前轮廓曲率特征值是否在预设轮廓曲率特征阈值区间内，所述轮廓曲率特征阈值区间为[轮廓曲率特征阈值最小值，轮廓曲率特征阈值最大值]；
若是，则判定所述肢体轮廓图像中无异物，否则，判定所述肢体轮廓图像中有异物。
在本实施例中，异物识别单元具体用于：
在所述肢体轮廓图像中标识出所述异物的位置；
根据所述异物在所述肢体轮廓图像中的位置，生成异物检测结果。
实施例七：
如图12所示，本发明一实施例提供的一种终端设备200，其包括：处理器210、存储器220以及存储在存储器220中并可在处理器210上运行的计算机程序，例如上述实施例一至三中的软件方法程序。处理器210执行计算机程序时实现上述各个人体图像映射方法实施例中的步骤，例如图1所示的步骤S110至S160。或者，处理器210执行计算机程序时实现上述各装置实施例中各单元的功能，例如图10所示单元110至160的功能。
示例性的，计算机程序可以被分割成一个或多个单元，一个或者多个单元被存储在存储器220中，并由处理器210执行，以完成本发明。一个或多个单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述计算机程序在终端设备200中的执行过程。例如，计算机程序可以被分割成人体灰度图像获取单元，人体轮廓图像提取单元，直方图构建单元，第一肢体位置获取单元和第二肢体位置获取单元，映射单元，各模块具体功能如下：
人体灰度图像获取单元，用于获取人体的毫米波灰度图像；
人体轮廓图像提取单元，用于提取所述毫米波灰度图像中的人体轮廓图像；
直方图构建单元，用于构建所述人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图；
第一肢体位置获取单元，用于根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置；
第二肢体位置获取单元，用于获取人体卡通图像中的人体的肢体位置；
映射单元，用于根据所述人体轮廓图像中的人体的肢体位置和所述人体卡通图像中的人体的肢体位置，构建所述人体轮廓图像和所述人体卡通图像中各肢体的位置映射关系。
终端设备200可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。终端设备200可包括，但不仅限于，处理器210、存储器220。本领域技术人员可以理解，图12仅仅是终端设备200的示例，并不构成对终端设备200的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如终端设备200还可以包括输入输出设备、网络接入设备、总线等。
所称处理器210可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器210、数字信号处理器210(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器210可以是微处理器210或者该处理器210也可以是任何常规的处理器210等。
存储器220可以是终端设备200的内部存储单元，例如终端设备200的硬盘或内存。存储器220也可以是终端设备200的外部存储设备，例如终端设备200上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)等。进一步地，存储器220还可以既包括终端设备200的内部存储单元也包括外部存储设备。存储器22061用于存储计算机程序以及终端设备200所需的其他程序和数据。存储器220还可以用于暂时地存储已经输出或者将要输出的数据。
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述系统中单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述或记载的部分，可以参见其它实施例的相关描述。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
在本发明所提供的实施例中，应该理解到，所揭露的装置/终端设备200和方法，可以通过其它的方式实现。例如，以上所描述的装置/终端设备200实施例仅仅是示意性的，例如，模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性，机械或其它的形式。
作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。 可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器210执行时，可实现上述各个方法实施例的步骤。。其中，计算机程序包括计算机程序代码，计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。计算机可读介质可以包括：能够携带计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器220、只读存储器220(ROM，Read-Only Memory)、随机存取存储器220(RAM，Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是，计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读介质不包括是电载波信号和电信信号。
以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围，均应包含在本发明的保护范围之内。
Claims (10)
- 一种人体图像映射方法，其特征在于，所述方法包括：获取人体的毫米波灰度图像；提取所述毫米波灰度图像中的人体轮廓图像；构建所述人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图；根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置；获取人体卡通图像中的人体的肢体位置；根据所述人体轮廓图像中的人体的肢体位置和所述人体卡通图像中的人体的肢体位置，构建所述人体轮廓图像和所述人体卡通图像中各肢体的位置映射关系。
- 如权利要求1所述的人体图像映射方法，其特征在于，根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置，包括：对所述垂直空间分布直方图进行局部最大值检测，获取人体的垂直中线横坐标；对所述水平空间分布直方图进行局部最小值检测，获取人体的手顶纵坐标、头顶纵坐标和脚底纵坐标；根据所述头顶纵坐标和所述脚底纵坐标，获取人体的身高；根据所述头顶纵坐标或所述脚底纵坐标中的任一个以及所述垂直中线横坐标、所述身高、所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取与人体的各肢体对应的肢体轮廓图像。
- 如权利要求2所述的人体图像映射方法，其特征在于，对所述水平空间分布直方图进行局部最小值检测，获取人体的手顶纵坐标、头顶纵坐标和脚底纵坐标，包括：对所述水平空间分布直方图的第一区域进行局部最小值检测，根据检测到的局部最小值所在的行，确定人体的手顶纵坐标；获取所述水平空间分布直方图的第二区域的每一行中像素值为第一预设像素阈值的像素点个数，根据所述像素点个数小于第一预设阈值的行，确定人体的头顶纵坐标；获取所述水平空间分布直方图的第三区域的每一行中像素值为第二预设像素阈值的像素点个数，根据所述像素点个数小于第二预设阈值的行，确定人体的脚底纵坐标。
- 如权利要求1所述的人体图像映射方法，其特征在于，所述方法还包括：根据预设异物特征识别模型，识别所述人体轮廓图像中的异物；根据所述人体轮廓图像中的人体的肢体位置，标识所述异物在所述人体轮廓图像上的位置；根据所述位置映射关系，标识所述异物在所述人体卡通图像上的位置并生成异物检测结果。
- 一种人体图像映射系统，其特征在于，所述系统包括：人体灰度图像获取单元，用于获取人体的毫米波灰度图像；人体轮廓图像提取单元，用于提取所述毫米波灰度图像中的人体轮廓图像；直方图构建单元，用于构建所述人体轮廓图像在垂直方向上的垂直空间分布直方图和在水平方向上的水平空间分布直方图；第一肢体位置获取单元，用于根据所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取所述人体轮廓图像中的人体的肢体位置；第二肢体位置获取单元，用于获取人体卡通图像中的人体的肢体位置；映射单元，用于根据所述人体轮廓图像中的人体的肢体位置和所述人体卡通图像中的人体的肢体位置，构建所述人体轮廓图像和所述人体卡通图像中各肢体的位置映射关系。
- 如权利要求5所述的人体图像映射系统，其特征在于，所述肢体位置获取单元包括：垂直中线坐标获取单元，用于对所述垂直空间分布直方图进行局部最大值检测，获取人体的垂直中线横坐标；手顶、头顶、脚底坐标获取单元，用于对所述水平空间分布直方图进行局部最小值检测，获取人体的手顶纵坐标、头顶纵坐标和脚底纵坐标；身高获取单元，用于根据所述头顶纵坐标和所述脚底纵坐标，获取人体的身高；肢体轮廓图像获取单元，用于根据所述头顶纵坐标或所述脚底纵坐标中的任一个以及所述垂直中线横坐标、所述身高、所述垂直空间分布直方图、所述水平空间分布直方图和预设人体比例模型，获取与人体的各肢体对应的肢体轮廓图像。
- 如权利要求6所述的人体图像映射系统，其特征在于，所述手顶、头顶、脚底坐标获取单元包括：手顶坐标获取单元，用于对所述水平空间分布直方图的第一区域进行局部最小值检测，根据检测到的局部最小值所在的行，确定人体的手顶纵坐标；头顶坐标获取单元，用于获取所述水平空间分布直方图的第二区域的每一行中像素值为第一预设像素阈值的像素点个数，根据所述像素点个数小于第一预设阈值的行，确定人体的头顶纵坐标；脚底坐标获取单元，用于获取所述水平空间分布直方图的第三区域的每一行中像素值为第二预设像素阈值的像素点个数，根据所述像素点个数小于第二预设阈值的行，确定人体的脚底纵坐标。
- 如权利要求5所述的人体图像映射系统，其特征在于，所述系统还包括：异物识别单元，用于根据预设异物特征识别模型，识别所述人体轮廓图像中的异物；第一异物标识单元，用于根据所述人体轮廓图像中的人体的肢体位置，标识所述异物在所述人体轮廓图像上的位置；第二异物标识单元，用于根据所述位置映射关系，标识所述异物在所述人体卡通图像上的位置并生成异物检测结果。
- 一种终端设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现如权利要求1至4任一项所述方法的步骤。
- 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1至4任一项所述方法的步骤。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710804373.8 | 2017-09-08 | ||
| CN201710804373.8A CN107730439B (zh) | 2017-09-08 | 2017-09-08 | 一种人体图像映射方法、系统及终端设备 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019047492A1 true WO2019047492A1 (zh) | 2019-03-14 |
Family
ID=61205888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/078038 WO2019047492A1 (zh) | 2017-09-08 | 2018-03-05 | 一种人体图像映射方法、系统及终端设备 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107730439B (zh) |
| WO (1) | WO2019047492A1 (zh) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111833420A (zh) * | 2020-07-07 | 2020-10-27 | 北京奇艺世纪科技有限公司 | 基于真人的图画自动生成方法、装置、系统及存储介质 |
| CN111950491A (zh) * | 2020-08-19 | 2020-11-17 | 成都飞英思特科技有限公司 | 一种人员密度的监控方法、装置及计算机可读存储介质 |
| CN113632139A (zh) * | 2019-03-28 | 2021-11-09 | 日本电气方案创新株式会社 | 部位检测设备、部位检测方法和计算机可读记录介质 |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107730439B (zh) * | 2017-09-08 | 2021-02-05 | 深圳市无牙太赫兹科技有限公司 | 一种人体图像映射方法、系统及终端设备 |
| CN109543582A (zh) * | 2018-11-15 | 2019-03-29 | 杭州芯影科技有限公司 | 基于毫米波图像的人体异物检测方法 |
| CN110334571B (zh) * | 2019-04-03 | 2022-12-20 | 复旦大学 | 一种基于卷积神经网络的毫米波图像人体隐私保护方法 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120194376A1 (en) * | 2007-12-18 | 2012-08-02 | Brijot Imaging Systems, Inc. | Millimeter Wave Energy Sensing Wand and Method |
| CN102708560A (zh) * | 2012-02-29 | 2012-10-03 | 北京无线电计量测试研究所 | 一种基于毫米波成像的隐私保护方法 |
| CN106127732A (zh) * | 2016-06-13 | 2016-11-16 | 深圳市太赫兹科技创新研究院 | 微波图像中人体性别检测方法和装置 |
| CN106372583A (zh) * | 2016-08-25 | 2017-02-01 | 华讯方舟科技有限公司 | 一种基于毫米波图像的人体异物检测方法及系统 |
| CN107730439A (zh) * | 2017-09-08 | 2018-02-23 | 深圳市无牙太赫兹科技有限公司 | 一种人体图像映射方法、系统及终端设备 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7961912B2 (en) * | 2004-10-14 | 2011-06-14 | Stevick Glen R | Method and apparatus for dynamic space-time imaging system |
| CN102608673A (zh) * | 2012-02-29 | 2012-07-25 | 北京无线电计量测试研究所 | 人体安检系统可疑物品图像显示方法 |
| CN104345351A (zh) * | 2013-07-23 | 2015-02-11 | 清华大学 | 人体安全检查的隐私保护方法和人体安全检查系统 |
| DE102013225283B4 (de) * | 2013-12-09 | 2023-04-27 | Rohde & Schwarz GmbH & Co. Kommanditgesellschaft | Verfahren und Vorrichtung zum Erfassen einer Rundumansicht |
| CN106443815B (zh) * | 2016-09-05 | 2019-01-25 | 华讯方舟科技有限公司 | 一种太赫兹安检方法及系统 |
-
2017
- 2017-09-08 CN CN201710804373.8A patent/CN107730439B/zh active Active
-
2018
- 2018-03-05 WO PCT/CN2018/078038 patent/WO2019047492A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120194376A1 (en) * | 2007-12-18 | 2012-08-02 | Brijot Imaging Systems, Inc. | Millimeter Wave Energy Sensing Wand and Method |
| CN102708560A (zh) * | 2012-02-29 | 2012-10-03 | 北京无线电计量测试研究所 | 一种基于毫米波成像的隐私保护方法 |
| CN106127732A (zh) * | 2016-06-13 | 2016-11-16 | 深圳市太赫兹科技创新研究院 | 微波图像中人体性别检测方法和装置 |
| CN106372583A (zh) * | 2016-08-25 | 2017-02-01 | 华讯方舟科技有限公司 | 一种基于毫米波图像的人体异物检测方法及系统 |
| CN107730439A (zh) * | 2017-09-08 | 2018-02-23 | 深圳市无牙太赫兹科技有限公司 | 一种人体图像映射方法、系统及终端设备 |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113632139A (zh) * | 2019-03-28 | 2021-11-09 | 日本电气方案创新株式会社 | 部位检测设备、部位检测方法和计算机可读记录介质 |
| CN111833420A (zh) * | 2020-07-07 | 2020-10-27 | 北京奇艺世纪科技有限公司 | 基于真人的图画自动生成方法、装置、系统及存储介质 |
| CN111833420B (zh) * | 2020-07-07 | 2023-06-30 | 北京奇艺世纪科技有限公司 | 基于真人的图画自动生成方法、装置、系统及存储介质 |
| CN111950491A (zh) * | 2020-08-19 | 2020-11-17 | 成都飞英思特科技有限公司 | 一种人员密度的监控方法、装置及计算机可读存储介质 |
| CN111950491B (zh) * | 2020-08-19 | 2024-04-02 | 成都飞英思特科技有限公司 | 一种人员密度的监控方法、装置及计算机可读存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107730439A (zh) | 2018-02-23 |
| CN107730439B (zh) | 2021-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019047492A1 (zh) | 一种人体图像映射方法、系统及终端设备 | |
| WO2018035814A1 (zh) | 一种基于毫米波图像的人体异物检测方法及系统 | |
| US10884116B2 (en) | Human-body foreign-matter detection method and system based on millimetre-wave image | |
| WO2017219391A1 (zh) | 一种基于三维数据的人脸识别系统 | |
| JP6507046B2 (ja) | 立体物検知装置及び立体物認証装置 | |
| CN102609683B (zh) | 一种基于单目视频的人体关节自动标注方法 | |
| US9723979B2 (en) | Method, apparatus and computer program product for positioning pupil | |
| CN106372583B (zh) | 一种基于毫米波图像的人体异物检测方法及系统 | |
| CN109446895B (zh) | 一种基于人体头部特征的行人识别方法 | |
| US20190213388A1 (en) | Intelligent Body Measurement | |
| WO2017215393A1 (zh) | 微波图像中人体性别检测方法和装置 | |
| CN110728196A (zh) | 一种人脸识别的方法、装置及终端设备 | |
| CN110782428B (zh) | 一种用于构建临床脑部ct图像roi模板的方法及系统 | |
| US20120114202A1 (en) | Image registration system and method for registering images for deformable surfaces | |
| WO2017070923A1 (zh) | 一种人脸识别方法和装置 | |
| Sagar et al. | Color channel based segmentation of skin lesion from clinical images for the detection of melanoma | |
| CN108108740B (zh) | 一种主动毫米波人体图像性别识别方法 | |
| CN111210423B (zh) | Nir图像的乳房轮廓提取方法、系统及装置 | |
| Arif et al. | Human pose estimation and object interaction for sports behaviour | |
| CN112907517A (zh) | 一种图像处理方法、装置、计算机设备及存储介质 | |
| Saha et al. | An approach to detect the region of interest of expressive face images | |
| Liu et al. | Automated binocular vision measurement of food dimensions and volume for dietary evaluation | |
| CN116189226A (zh) | 一种基于毫米波安检图像危险品坐标映射方法 | |
| Nahari et al. | Cow weight estimation using local adaptive thresholding method and connected component labelling | |
| Costa et al. | Tessellation-based coarse registration method for 3d reconstruction of the female torso |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18853576 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18853576 Country of ref document: EP Kind code of ref document: A1 |