CN117115856A - Target detection method based on image fusion, human body security inspection equipment and storage medium - Google Patents
Target detection method based on image fusion, human body security inspection equipment and storage medium Download PDFInfo
- Publication number
- CN117115856A CN117115856A CN202310968561.XA CN202310968561A CN117115856A CN 117115856 A CN117115856 A CN 117115856A CN 202310968561 A CN202310968561 A CN 202310968561A CN 117115856 A CN117115856 A CN 117115856A
- Authority
- CN
- China
- Prior art keywords
- rectangular frame
- millimeter wave
- visible light
- coordinate
- article
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 135
- 238000007689 inspection Methods 0.000 title claims abstract description 42
- 230000004927 fusion Effects 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000013507 mapping Methods 0.000 claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 29
- 238000012545 processing Methods 0.000 claims description 24
- 238000003384 imaging method Methods 0.000 claims description 16
- 238000004891 communication Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 4
- 238000012790 confirmation Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The application discloses a target detection method, human body security inspection equipment and medium based on image fusion, wherein the method comprises the steps of obtaining visible light human body rectangular frame coordinates and visible light article rectangular frame coordinates corresponding to images, and obtaining millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates corresponding to millimeter wave images; calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing a coordinate scaling pair Ji Canshu to obtain millimeter wave article alignment coordinates; calculating the intersection ratio of the relative rectangular frame coordinates between the visible light article rectangular frame coordinates and the visible light human body rectangular frame coordinates and the millimeter wave article alignment coordinates, and mapping the millimeter wave article rectangular frame coordinates in the visible light image according to the intersection ratio. According to the application, the visible light image is utilized to assist in the article detection of the millimeter wave image, and compared with the scheme of carrying out the article detection based on the millimeter wave image, the false alarm rate of the article detection is reduced.
Description
Technical Field
The application relates to the technical field of millimeter wave image processing, in particular to a target detection method based on image fusion, human body security inspection equipment and a storage medium.
Background
In order to meet the high-frame-rate real-time imaging requirement, the current real-time millimeter wave imaging human body security inspection equipment has great sacrifice on imaging performance, and the image resolution and the definition of human bodies and articles in the image are not ideal. In the human body security check scene of the rail transit industry with high human flow, when passengers carry plastic bags or handbags or umbrellas and the like filled with articles to pass through the human body security check equipment, the passengers can hold the plastic bags or handbags or umbrellas on the hands or carry the plastic bags or the umbrellas on the shoulders, and other carrying modes pass through the security check equipment.
Disclosure of Invention
The embodiment of the application provides a target detection method, human body security inspection equipment and a storage medium based on image fusion, which can effectively reduce the article detection false alarm rate of human body security inspection products, thereby improving the customer experience of the human body security inspection products.
In a first aspect, an embodiment of the present application provides an image fusion-based target detection method, which is applied to a human body security inspection device, and the method includes:
obtaining a visible light image and a millimeter wave image corresponding to a reference object;
performing image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light object rectangular frame coordinate;
performing image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates;
calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates;
calculating the intersection ratio between a first relative rectangular frame coordinate and the alignment coordinate of the millimeter wave article, and mapping the rectangular frame coordinate of the millimeter wave article in the visible light image according to the intersection ratio, wherein the first relative rectangular frame coordinate is the relative rectangular frame coordinate between the rectangular frame coordinate of the visible light article and the rectangular frame coordinate of the visible light human body.
In some embodiments, the human body security inspection device includes a millimeter wave imaging device and a visible light camera, and in the case that the millimeter wave image is a millimeter wave video stream and the visible light image is a visible light video stream, before the obtaining the visible light image and the millimeter wave image corresponding to the reference object, the method further includes:
adjusting shooting direction information and mounting position information of the visible light camera according to the position information of the millimeter wave imaging device so that the visual angle corresponding to the visible light image and the millimeter wave image is the same;
and determining a reference frame rate, wherein the reference frame rate is the frame rate of the millimeter wave video stream, and adjusting the frame rate of the visible light camera according to the reference frame rate so that the frame rate of the visible light camera is the same as the reference frame rate.
In some embodiments, the scaling calculation is performed on the coordinates of the rectangular frame of the millimeter wave article by using the coordinate scaling alignment parameter to obtain millimeter wave article alignment coordinates, including:
calculating a second relative rectangular frame coordinate between the millimeter wave article rectangular frame coordinate and the millimeter wave human body rectangular frame coordinate;
and scaling calculation is carried out on the coordinates of the second relative rectangular frame by utilizing the coordinate scaling alignment parameters, so that the millimeter wave article alignment coordinates are obtained.
In some embodiments, in the case where the first relative rectangular frame coordinates include at least a first coordinate and a second coordinate, the intersection ratio includes at least a first intersection ratio between the first coordinate and the millimeter wave article alignment coordinate and a second intersection ratio between the first coordinate and the millimeter wave article alignment coordinate, the mapping the millimeter wave article rectangular frame coordinates in the visible light image according to the intersection ratio includes:
determining a target intersection ratio value from the first intersection ratio value and the second intersection ratio value, wherein the target intersection ratio value is the intersection ratio value with the largest numerical value in the first intersection ratio value and the second intersection ratio value, and the target intersection ratio value is larger than a preset threshold value;
determining the rectangular frame coordinates of the object visible light object corresponding to the object intersection ratio;
and mapping the millimeter wave article rectangular frame coordinates on the target visible light article rectangular frame coordinates of the visible light image.
In some embodiments, the human body security inspection device is communicatively connected to a terminal, the visible light detection result further includes item category information corresponding to the visible light item rectangular frame coordinates, and after the mapping the millimeter wave item rectangular frame coordinates in the visible light image according to the intersection ratio, the method further includes:
carrying out false alarm detection on the article category information according to a preset knowledge graph to obtain a false alarm detection result;
and when the false alarm detection result does not meet the false alarm condition, generating alarm prompt information according to the false alarm detection result and the article category information, and sending the alarm prompt information to the terminal.
In some embodiments, the millimeter wave article alignment coordinates are derived according to the following formula:
wherein, (Sx, sy) is the coordinate scaling alignment parameter, (RDx, RDy) is the upper left corner pixel point coordinate corresponding to the second relative rectangular frame coordinate, (RDw, RDh) is the width value and the height value of the rectangular frame corresponding to the second relative rectangular frame coordinate, (Dx, dy) is the upper left corner pixel point coordinate corresponding to the millimeter wave article rectangular frame coordinate, and (Bx, by) is the upper left corner pixel point coordinate corresponding to the millimeter wave human body rectangular frame coordinate.
In some embodiments, the target visible light item rectangular box coordinates are derived according to the following formula:
V(x,y,w,h)＝(RC x +A x ,RC y +A y ,w,h)；
wherein V (x, y, w, h) is the rectangular frame coordinate of the target visible light article, (RCx, RCy) is the upper left corner pixel point coordinate corresponding to the first relative rectangular frame coordinate, and (Ax, ay) is the upper left corner pixel point coordinate corresponding to the visible light human body rectangular frame coordinate.
In a second aspect, an embodiment of the present application provides a human body security inspection apparatus, including:
the image acquisition module is used for acquiring a visible light image and a millimeter wave image corresponding to the reference object;
the visible light detection module is used for carrying out image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light article rectangular frame coordinate;
the millimeter wave detection module is used for carrying out image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates;
the millimeter wave article alignment coordinate acquisition module is used for calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates;
and the image mapping module is used for calculating the intersection ratio between the first relative rectangular frame coordinate and the alignment coordinate of the millimeter wave article and mapping the rectangular frame coordinate of the millimeter wave article in the visible light image according to the intersection ratio, wherein the first relative rectangular frame coordinate is the relative rectangular frame coordinate between the rectangular frame coordinate of the visible light article and the rectangular frame coordinate of the visible light human body.
In a third aspect, an embodiment of the present application further provides a human body security inspection device, including at least one control processor and a memory communicatively connected to the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform the image fusion-based object detection method according to the first aspect.
In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium storing computer executable instructions for performing the image fusion-based object detection method according to the first aspect.
The embodiment of the application provides a target detection method, human body security inspection equipment and a medium based on image fusion, wherein the method comprises the steps of obtaining a visible light image and a millimeter wave image corresponding to a reference object; performing image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light object rectangular frame coordinate; performing image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates; calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates; calculating the intersection ratio between a first relative rectangular frame coordinate and the alignment coordinate of the millimeter wave article, and mapping the rectangular frame coordinate of the millimeter wave article in the visible light image according to the intersection ratio, wherein the first relative rectangular frame coordinate is the relative rectangular frame coordinate between the rectangular frame coordinate of the visible light article and the rectangular frame coordinate of the visible light human body. According to the embodiment of the application, the visible light image is utilized to assist the article detection and confirmation process of the millimeter wave image, and compared with the conventional scheme of carrying out article detection based on the millimeter wave image, the article detection false alarm rate of the human body security inspection product can be effectively reduced, so that the customer experience of the human body security inspection product is improved.
Drawings
FIG. 1 is a flow chart of steps of an image fusion-based object detection method according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating steps for adjusting parameters of a camera of a security inspection device according to another embodiment of the present application;
fig. 3 is a flowchart showing steps for obtaining alignment coordinates of a millimetric-wave article according to another embodiment of the present application;
fig. 4 is a flowchart showing a step of mapping coordinates of a rectangular frame of a millimetric wave article on a visible light image according to another embodiment of the present application;
FIG. 5 is a flowchart illustrating steps for false alarm detection of item class information according to another embodiment of the present application;
FIG. 6 is a schematic block diagram of a human body security inspection apparatus according to another embodiment of the present application;
fig. 7 is a block diagram of a human body security inspection apparatus according to another embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
It will be appreciated that although functional block diagrams are depicted in the device diagrams, logical sequences are shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the block diagrams in the device. The terms first, second and the like in the description, in the claims and in the above-described figures, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order.
At present, in order to meet the requirement of high-frame-rate real-time imaging, the current real-time millimeter wave imaging human body security inspection equipment has great sacrifice on imaging performance, and the image resolution and the definition of human bodies and articles in images are not ideal. In the human body security check scene of the rail transit industry with high human flow, when passengers carry plastic bags or handbags or umbrellas and the like filled with articles to pass through the human body security check equipment, the passengers can hold the plastic bags or handbags or umbrellas on the hands or carry the plastic bags or the umbrellas on the shoulders, and other carrying modes pass through the security check equipment.
In order to solve the problems, the embodiment of the application provides a target detection method, a human body security inspection device and a medium based on image fusion, wherein the method comprises the steps of obtaining a visible light image and a millimeter wave image corresponding to a reference object; performing image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light object rectangular frame coordinate; performing image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates; calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates; calculating the intersection ratio between a first relative rectangular frame coordinate and the alignment coordinate of the millimeter wave article, and mapping the rectangular frame coordinate of the millimeter wave article in the visible light image according to the intersection ratio, wherein the first relative rectangular frame coordinate is the relative rectangular frame coordinate between the rectangular frame coordinate of the visible light article and the rectangular frame coordinate of the visible light human body. According to the embodiment of the application, the visible light image is utilized to assist the article detection and confirmation process of the millimeter wave image, and compared with the conventional scheme of carrying out article detection based on the millimeter wave image, the article detection false alarm rate of the human body security inspection product can be effectively reduced, so that the customer experience of the human body security inspection product is improved.
Embodiments of the present application will be further described below with reference to the accompanying drawings.
As shown in fig. 1, fig. 1 is a flowchart illustrating steps of an image fusion-based object detection method according to an embodiment of the present application, and the embodiment of the present application provides an image fusion-based object detection method, which is applied to a human body security inspection device, and includes, but is not limited to, the following steps:
step S110, obtaining a visible light image and a millimeter wave image corresponding to a reference object;
it should be noted that, the embodiment of the present application does not limit the specific manner of acquiring the visible light image and the millimeter wave image corresponding to the reference object, and may be that when the human body security inspection device senses the person to be inspected, that is, the reference object passes through the human body security inspection device, the corresponding image acquisition module in the human body security inspection device is triggered to acquire the visible light image and the millimeter wave image corresponding to the reference object; the method can also be used for selecting the target visible light image and the target millimeter wave image associated with the reference object from the visible light image and the millimeter wave image obtained in real time, and the method can be selected by a person skilled in the art according to actual conditions.
It is understood that acquiring the visible light image and the millimeter wave image corresponding to the reference object can provide an effective data basis for the subsequent target detection process.
Step S120, performing image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light article rectangular frame coordinate;
step S130, performing image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates;
it should be noted that, the embodiment of the present application is not limited to the specific first target detection algorithm and the specific second target detection algorithm, and may be a yolov5 algorithm or a yolov7 algorithm, etc., which may be selected by a person skilled in the art according to actual situations.
In addition, before performing image processing on the visible light image by using the first object detection algorithm, the method of the present embodiment may further include: obtaining a visible light image set carrying priori information marks, and carrying out model training on a first initial model according to the visible light image set carrying priori information marks and a yolov5 algorithm until the first initial model converges to obtain a first target detection model, namely a model corresponding to the first target detection algorithm; likewise, before performing image processing on the millimeter wave image using the second object detection algorithm, the method of the present embodiment may further include: and acquiring a millimeter wave image set carrying priori information marks, and performing model training on the second initial model according to the millimeter wave image set carrying priori information marks and the yolov5 algorithm until the second initial model converges to obtain a second target detection model, namely a model corresponding to the second target detection algorithm.
It can be understood that performing image processing on the visible light image according to the first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light object rectangular frame coordinate; and performing image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates, and an effective data basis can be provided for subsequently acquiring millimeter wave article alignment coordinates.
Step S140, calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates;
it should be noted that, assuming that the width and height of the visible light human rectangular frame coordinates are a (w, h), the width and height of the millimeter wave human rectangular frame coordinates are B (w, h), and taking a (w, h) as a reference, the coordinate scaling pair Ji Canshu (Sx, sy) of the millimeter wave human rectangular frame coordinates with respect to the visible light human rectangular frame coordinates is calculated as follows:
and step S150, calculating the intersection ratio between the first relative rectangular frame coordinates and the alignment coordinates of the millimeter wave article, and mapping the rectangular frame coordinates of the millimeter wave article in the visible light image according to the intersection ratio, wherein the first relative rectangular frame coordinates are the relative rectangular frame coordinates between the rectangular frame coordinates of the visible light article and the rectangular frame coordinates of the visible light human body.
It can be understood that the existing millimeter wave imaging device is limited by performance limitation, the acquired millimeter wave image is deficient in human body, and only based on the millimeter wave image, key points of human bones cannot be extracted correctly by utilizing a deep learning posture estimation algorithm, so that the coordinate alignment of the millimeter wave and the same object in the visible light image cannot be realized, and the accuracy of the existing target detection result is low.
It can be understood that, calculate the coordinate scaling alignment parameter of millimeter wave human body rectangular frame coordinate relative to visible light human body rectangular frame coordinate, utilize coordinate scaling alignment parameter to carry out scaling calculation to millimeter wave article rectangular frame coordinate, obtain millimeter wave article alignment coordinate, calculate the ratio of crossing between first relative rectangular frame coordinate and the millimeter wave article alignment coordinate, and map millimeter wave article rectangular frame coordinate in visible light image according to the ratio of crossing, utilize visible light image to assist millimeter wave image's article detection confirmation process, compare in the scheme that only carries out article detection based on millimeter wave image at present, can effectively reduce article detection false alarm rate, thereby promote human body security inspection product's customer experience sense.
In addition, in some embodiments, the human body security inspection device includes a millimeter wave imaging device and a visible light camera, and in the case that the millimeter wave image is a millimeter wave video stream, referring to fig. 2, and the visible light image is a visible light video stream, before executing step S110 of fig. 1, the image fusion-based object detection method of the present embodiment includes, but is not limited to, the following steps:
step S210, adjusting shooting direction information and installation position information of a visible light camera according to position information of a millimeter wave imaging device so that the visual angles of a visible light image and a millimeter wave image are the same;
step S220, determining a reference frame rate, wherein the reference frame rate is the frame rate of the millimeter wave video stream, and adjusting the frame rate of the visible light camera according to the reference frame rate so that the frame rate of the visible light camera is the same as the reference frame rate.
It can be understood that before the visible light image and the millimeter wave image corresponding to the reference object are acquired through the human body security inspection equipment, the shooting direction information and the installation position information of the visible light camera are adjusted according to the position information of the millimeter wave imaging device of the human body security inspection equipment, so that the visible light image acquired by the visible light camera is the same as the visual angle corresponding to the millimeter wave image acquired by the millimeter wave imaging device; and the frame rate of the visible light camera is adjusted according to the reference frame rate, so that the frame rate of the visible light camera is the same as the reference frame rate, the same position and the same posture of a human body at the same moment can be recorded in the same frame image acquired by the visible light camera and the millimeter wave imaging device, and an effective data basis can be provided for the subsequent coordinate alignment operation of the same object in the visible light image and the millimeter wave image.
Additionally, in some embodiments, referring to FIG. 3, step S140 of the FIG. 1 embodiment includes, but is not limited to, the steps of:
step S310, calculating second relative rectangular frame coordinates between the rectangular frame coordinates of the millimeter wave article and the rectangular frame coordinates of the millimeter wave human body;
and step S320, scaling calculation is carried out on the coordinates of the second relative rectangular frame by utilizing the coordinate scaling alignment parameters, so that the millimeter wave article alignment coordinates are obtained.
Note that the expression corresponding to the millimeter wave article alignment coordinates is as follows:
wherein, (Sx, sy) is the coordinate scaling alignment parameter, (RDx, RDy) is the upper left corner pixel point coordinate corresponding to the second relative rectangular frame coordinate, (RDw, RDh) is the width value and the height value of the rectangular frame corresponding to the second relative rectangular frame coordinate, (Dx, dy) is the upper left corner pixel point coordinate corresponding to the millimeter wave article rectangular frame coordinate, and (Bx, by) is the upper left corner pixel point coordinate corresponding to the millimeter wave human body rectangular frame coordinate.
In addition, in some embodiments, in the case where the first relative rectangular frame coordinates include at least a first coordinate and a second coordinate, the intersection ratio includes at least a first intersection ratio and a second intersection ratio, the first intersection ratio is an intersection ratio between the first coordinate and the millimeter wave article alignment coordinate, and the second intersection ratio is an intersection ratio between the first coordinate and the millimeter wave article alignment coordinate, referring to fig. 4, step S150 of the embodiment of fig. 1 includes, but is not limited to, the steps of:
step S410, determining a target cross ratio from the first cross ratio and the second cross ratio, wherein the target cross ratio is the cross ratio with the largest value in the first cross ratio and the second cross ratio, and the target cross ratio is larger than a preset threshold;
step S420, determining the rectangular frame coordinates of the object visible light object corresponding to the object intersection ratio;
step S430, the millimeter wave article rectangular frame coordinates are mapped on the target visible light article rectangular frame coordinates of the visible light image.
It may be understood that, assuming that, in the millimeter wave detection result, the number of rectangular frame coordinates of the millimeter wave article is 1, where the first relative rectangular frame coordinates include at least a first coordinate and a second coordinate, this time indicates that 2 articles are detected in the same frame of visible light image corresponding to the millimeter wave image, the first coordinate is RC1 (x, y, w, h), the second coordinate is RC2 (x, y, w, h), where RC1 (w, h) and RC2 (w, h) are respectively the width value and the height value of the rectangular frame corresponding to the two first relative rectangular frame coordinates, and RC1 (x, y) and RC2 (x, y) are respectively the upper left corner pixel point coordinates corresponding to the two first relative rectangular frame coordinates. The first and second cross ratios are calculated as follows:
at this time, a target intersection ratio is determined from the first intersection ratio IOU (M, RC 1) and the second intersection ratio IOU (M, RC 2), wherein the target intersection ratio is calculated as follows:
V(x,y,w,h)＝(RC x +A x ,RC y +A y ,w,h)；
wherein V (x, y, w, h) is the rectangular frame coordinate of the target visible light article, (RCx, RCy) is the upper left corner pixel point coordinate corresponding to the first relative rectangular frame coordinate, and (Ax, ay) is the upper left corner pixel point coordinate corresponding to the visible light human body rectangular frame coordinate. Based on the above embodiment, 2 objects are detected in the visible light image, and each object corresponds to the case where the first coordinate is RC1 (x, y, w, h) and the second coordinate is RC2 (x, y, w, h), and the target overlap ratio needs to satisfy the following screening conditions: the target cross ratio is the cross ratio with the largest value in the first cross ratio and the second cross ratio, and the target cross ratio is larger than a preset threshold, and the specific screening steps are as follows:
if IOU (M, RC 1) > IOU (M, RC 2) and IOU (M, RC 1) > thresh, then IOU (M, RC 1) is the target intersection ratio, then the corresponding target visible article rectangular frame coordinates are expressed as:
V(x,y,w,h)＝(RC 1 x +A x ,RC 1 y +A y ,w,h)；
at this time, it is explained that the degree of alignment of the article rectangular frame corresponding to the first coordinate and the millimeter wave article rectangular frame is highest, that is, the likelihood that the article corresponding to the first coordinate is the same article as the article corresponding to the millimeter wave article rectangular frame is higher.
If IOU (M, RC 2) > IOU (M, RC 1) and IOU (M, RC 2) > thresh, then IOU (M, RC 2) is the target intersection ratio, then the corresponding target visible article rectangular frame coordinates are expressed as:
V(x,y,w,h)＝(RC 2 x +A x ,RC 2 y +A y ,w,h)；
at this time, it is explained that the degree of alignment of the article rectangular frame corresponding to the second coordinate and the millimeter wave article rectangular frame is highest, that is, the likelihood that the article corresponding to the second coordinate is the same article as the article corresponding to the millimeter wave article rectangular frame is higher.
Wherein thresh represents a preset threshold value, and the value range is [0.0,1.0].
After the target visible light article rectangular frame coordinates are determined, the millimeter wave article rectangular frame coordinates D (x, y, w, h) are mapped on the target visible light article rectangular frame coordinates V (x, y, w, h) of the visible light image to achieve the same article coordinate alignment in different images.
In addition, in some embodiments, the human body security inspection device is in communication connection with the terminal, the visible light detection result further includes item category information corresponding to the coordinates of the rectangular frame of the visible light item, and after executing step S150 in the embodiment of fig. 1, the image fusion-based object detection method in the embodiment of the present application includes, but is not limited to, the following steps:
step S510, carrying out false alarm detection on the article category information according to a preset knowledge graph to obtain a false alarm detection result;
and step S520, when the false alarm detection result does not meet the false alarm condition, generating alarm prompt information according to the false alarm detection result and the article category information, and sending the alarm prompt information to the terminal.
It may be understood that the visible light detection result may further include item category information corresponding to the visible light item rectangular frame coordinates, after executing step S150 in the embodiment of fig. 1, namely, mapping the millimeter wave item rectangular frame coordinates on the visible light image, performing false alarm detection on the item category information according to a preset knowledge graph to obtain a false alarm detection result, and determining that the false alarm detection result meets the false alarm condition when the false alarm detection result indicates that the item category information is a mobile phone, an umbrella, a cup, a hat, or the like; when the false alarm detection result indicates that the article type information is a plastic bag or a handbag, and the like, the false alarm detection result is determined to not meet the false alarm condition, the corresponding article is determined to be a suspected dangerous article, alarm prompt information is generated according to the false alarm detection result and the article type information, and the alarm prompt information is sent to the terminal, so that security personnel corresponding to the terminal can carry out manual recheck.
In addition, referring to fig. 6, fig. 6 is a schematic block diagram of a human body security inspection apparatus according to another embodiment of the present application, and one embodiment of the present application further provides a human body security inspection apparatus 600, where the human body security inspection apparatus 600 includes:
an image acquisition module 610, configured to acquire a visible light image and a millimeter wave image corresponding to a reference object;
the visible light detection module 620 is configured to perform image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, where the visible light detection result includes a visible light human rectangular frame coordinate and a visible light object rectangular frame coordinate;
the millimeter wave detection module 630 is configured to perform image processing on the millimeter wave image according to a second target detection algorithm, so as to obtain a millimeter wave detection result corresponding to the reference object, where the millimeter wave detection result includes a millimeter wave human rectangular frame coordinate and a millimeter wave article rectangular frame coordinate;
the millimeter wave article alignment coordinate acquisition module 640 is used for calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates;
the image mapping module 650 is configured to calculate an intersection ratio between the first relative rectangular frame coordinate and the alignment coordinate of the millimeter-wave article, and map the rectangular frame coordinate of the millimeter-wave article in the visible light image according to the intersection ratio, where the first relative rectangular frame coordinate is a relative rectangular frame coordinate between the rectangular frame coordinate of the visible light article and the rectangular frame coordinate of the visible light human body.
It should be noted that, the specific embodiment of the human body security inspection apparatus 600 is substantially the same as the specific embodiment and the specific step principle of the above-mentioned target detection method based on image fusion, and will not be described herein again.
As shown in fig. 7, fig. 7 is a structural diagram of a human body security inspection apparatus according to an embodiment of the present application. The application also provides a human body security inspection device 700, comprising:
the processor 710 may be implemented by a general purpose central processing unit (Central Processing Unit, CPU), a microprocessor, an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, etc. for executing related programs to implement the technical scheme provided by the embodiments of the present application;
the Memory 720 may be implemented in the form of a Read Only Memory (ROM), a static storage device, a dynamic storage device, or a random access Memory (Random Access Memory, RAM). The memory 720 may store an operating system and other application programs, and when the technical solution provided in the embodiments of the present specification is implemented by software or firmware, relevant program codes are stored in the memory 720 and the processor 710 invokes an image fusion-based object detection method for performing the embodiments of the present application, for example, performing the above-described method steps S110 to S150 in fig. 1, the method steps S210 to S220 in fig. 2, the method steps S310 to S320 in fig. 3, the method steps S410 to S430 in fig. 4, and the method steps S510 to S520 in fig. 5;
an input/output interface 730 for implementing information input and output;
the communication interface 740 is configured to implement communication interaction between the device and other devices, and may implement communication in a wired manner (e.g., USB, network cable, etc.), or may implement communication in a wireless manner (e.g., mobile network, WIFI, bluetooth, etc.);
bus 750 transfers information between the various components of the device (e.g., processor 710, memory 720, input/output interface 730, and communication interface 740);
wherein processor 710, memory 720, input/output interface 730, and communication interface 740 implement a communication connection among each other within the device via bus 750.
The embodiment of the present application also provides a storage medium, which is a computer-readable storage medium storing a computer program that when executed by a processor implements the above-described image fusion-based object detection method, for example, performs the above-described method steps S110 to S150 in fig. 1, the above-described method steps S210 to S220 in fig. 2, the method steps S310 to S320 in fig. 3, the method steps S410 to S430 in fig. 4, and the method steps S510 to S520 in fig. 5.
The memory, as a non-transitory computer readable storage medium, may be used to store non-transitory software programs as well as non-transitory computer executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory remotely located relative to the processor, the remote memory being connectable to the processor through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof. The apparatus embodiments described above are merely illustrative, in which the elements illustrated as separate components may or may not be physically separate, implemented to reside in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically include computer readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and may include any information delivery media.
While the preferred embodiment of the present application has been described in detail, the present application is not limited to the above embodiments, and those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit and scope of the present application, and these equivalent modifications or substitutions are included in the scope of the present application as defined in the appended claims.
Claims (10)
1. An object detection method based on image fusion, which is characterized by being applied to human body security inspection equipment, comprises the following steps:
obtaining a visible light image and a millimeter wave image corresponding to a reference object;
performing image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light object rectangular frame coordinate;
performing image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates;
calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates;
calculating the intersection ratio between a first relative rectangular frame coordinate and the alignment coordinate of the millimeter wave article, and mapping the rectangular frame coordinate of the millimeter wave article in the visible light image according to the intersection ratio, wherein the first relative rectangular frame coordinate is the relative rectangular frame coordinate between the rectangular frame coordinate of the visible light article and the rectangular frame coordinate of the visible light human body.
2. The image fusion-based target detection method according to claim 1, wherein the human body security inspection device includes a millimeter wave imaging device and a visible light camera, and when the millimeter wave image is a millimeter wave video stream and the visible light image is a visible light video stream, before the obtaining the visible light image and the millimeter wave image corresponding to the reference object, the method further includes:
adjusting shooting direction information and mounting position information of the visible light camera according to the position information of the millimeter wave imaging device so that the visual angle corresponding to the visible light image and the millimeter wave image is the same;
and determining a reference frame rate, wherein the reference frame rate is the frame rate of the millimeter wave video stream, and adjusting the frame rate of the visible light camera according to the reference frame rate so that the frame rate of the visible light camera is the same as the reference frame rate.
3. The image fusion-based object detection method according to claim 1, wherein the scaling calculation is performed on the millimeter wave article rectangular frame coordinates by using the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates, and the method comprises:
calculating a second relative rectangular frame coordinate between the millimeter wave article rectangular frame coordinate and the millimeter wave human body rectangular frame coordinate;
and scaling calculation is carried out on the coordinates of the second relative rectangular frame by utilizing the coordinate scaling alignment parameters, so that the millimeter wave article alignment coordinates are obtained.
4. The image fusion-based object detection method according to claim 2, wherein in a case where the first relative rectangular frame coordinates include at least a first coordinate and a second coordinate, the intersection ratio includes at least a first intersection ratio, which is an intersection ratio between the first coordinate and the millimeter wave article alignment coordinate, and a second intersection ratio, which is an intersection ratio between the first coordinate and the millimeter wave article alignment coordinate, the mapping the millimeter wave article rectangular frame coordinates in the visible light image according to the intersection ratio, includes:
determining a target intersection ratio value from the first intersection ratio value and the second intersection ratio value, wherein the target intersection ratio value is the intersection ratio value with the largest numerical value in the first intersection ratio value and the second intersection ratio value, and the target intersection ratio value is larger than a preset threshold value;
determining the rectangular frame coordinates of the object visible light object corresponding to the object intersection ratio;
and mapping the millimeter wave article rectangular frame coordinates on the target visible light article rectangular frame coordinates of the visible light image.
5. The image fusion-based object detection method according to claim 1, wherein the human body security inspection device is in communication connection with a terminal, the visible light detection result further includes item category information corresponding to the visible light item rectangular frame coordinates, and after the mapping of the millimeter wave item rectangular frame coordinates in the visible light image according to the intersection ratio, the method further includes:
carrying out false alarm detection on the article category information according to a preset knowledge graph to obtain a false alarm detection result;
and when the false alarm detection result does not meet the false alarm condition, generating alarm prompt information according to the false alarm detection result and the article category information, and sending the alarm prompt information to the terminal.
6. The image fusion-based object detection method according to claim 3, wherein the millimetric wave article alignment coordinates are obtained according to the following formula:
wherein, (Sx, sy) is the coordinate scaling alignment parameter, (RDx, RDy) is the upper left corner pixel point coordinate corresponding to the second relative rectangular frame coordinate, (RDw, RDh) is the width value and the height value of the rectangular frame corresponding to the second relative rectangular frame coordinate, (Dx, dy) is the upper left corner pixel point coordinate corresponding to the millimeter wave article rectangular frame coordinate, and (Bx, by) is the upper left corner pixel point coordinate corresponding to the millimeter wave human body rectangular frame coordinate.
7. The image fusion-based object detection method according to claim 4, wherein the coordinates of the rectangular frame of the object visible light object are obtained according to the following formula:
V(x,y,w,h)＝(RC x +A x ,RC y +A y ,w,h)；
wherein V (x, y, w, h) is the rectangular frame coordinate of the target visible light article, (RCx, RCy) is the upper left corner pixel point coordinate corresponding to the first relative rectangular frame coordinate, and (Ax, ay) is the upper left corner pixel point coordinate corresponding to the visible light human body rectangular frame coordinate.
8. A human body security inspection device, comprising:
the image acquisition module is used for acquiring a visible light image and a millimeter wave image corresponding to the reference object;
the visible light detection module is used for carrying out image processing on the visible light image according to a first target detection algorithm to obtain a visible light detection result corresponding to the reference object, wherein the visible light detection result comprises a visible light human body rectangular frame coordinate and a visible light article rectangular frame coordinate;
the millimeter wave detection module is used for carrying out image processing on the millimeter wave image according to a second target detection algorithm to obtain a millimeter wave detection result corresponding to the reference object, wherein the millimeter wave detection result comprises millimeter wave human body rectangular frame coordinates and millimeter wave article rectangular frame coordinates;
the millimeter wave article alignment coordinate acquisition module is used for calculating coordinate scaling alignment parameters of the millimeter wave human body rectangular frame coordinates relative to the visible light human body rectangular frame coordinates, and scaling the millimeter wave article rectangular frame coordinates by utilizing the coordinate scaling alignment parameters to obtain millimeter wave article alignment coordinates;
and the image mapping module is used for calculating the intersection ratio between the first relative rectangular frame coordinate and the alignment coordinate of the millimeter wave article and mapping the rectangular frame coordinate of the millimeter wave article in the visible light image according to the intersection ratio, wherein the first relative rectangular frame coordinate is the relative rectangular frame coordinate between the rectangular frame coordinate of the visible light article and the rectangular frame coordinate of the visible light human body.
9. A human body security device comprising at least one control processor and a memory for communication connection with the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform the image fusion-based object detection method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing computer-executable instructions for causing a computer to perform the image fusion-based object detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310968561.XA CN117115856B (en) | 2023-08-02 | 2023-08-02 | Target detection method based on image fusion, human body security inspection equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310968561.XA CN117115856B (en) | 2023-08-02 | 2023-08-02 | Target detection method based on image fusion, human body security inspection equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117115856A true CN117115856A (en) | 2023-11-24 |
| CN117115856B CN117115856B (en) | 2024-04-05 |
Family
ID=88803009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310968561.XA Active CN117115856B (en) | 2023-08-02 | 2023-08-02 | Target detection method based on image fusion, human body security inspection equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117115856B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020206861A1 (en) * | 2019-04-08 | 2020-10-15 | 江西理工大学 | Yolo v3-based detection method for key object at transportation junction |
| WO2023001063A1 (en) * | 2021-07-19 | 2023-01-26 | 北京鹰瞳科技发展股份有限公司 | Target detection method and apparatus, electronic device, and storage medium |
| CN116051612A (en) * | 2023-01-17 | 2023-05-02 | 厦门瑞为信息技术有限公司 | Doll mapping method, device, medium and equipment for suspicious articles |
| CN116453160A (en) * | 2023-04-07 | 2023-07-18 | 北京无线电计量测试研究所 | Suspicious article non-stop detection method and system based on millimeter wave imaging |
-
2023
- 2023-08-02 CN CN202310968561.XA patent/CN117115856B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020206861A1 (en) * | 2019-04-08 | 2020-10-15 | 江西理工大学 | Yolo v3-based detection method for key object at transportation junction |
| WO2023001063A1 (en) * | 2021-07-19 | 2023-01-26 | 北京鹰瞳科技发展股份有限公司 | Target detection method and apparatus, electronic device, and storage medium |
| CN116051612A (en) * | 2023-01-17 | 2023-05-02 | 厦门瑞为信息技术有限公司 | Doll mapping method, device, medium and equipment for suspicious articles |
| CN116453160A (en) * | 2023-04-07 | 2023-07-18 | 北京无线电计量测试研究所 | Suspicious article non-stop detection method and system based on millimeter wave imaging |
Non-Patent Citations (1)
| Title |
|---|
| 陆峰;徐友春;李永乐;王任栋;王东敏;: "基于多传感器数据融合的障碍物检测与跟踪", 军事交通学院学报, no. 02, 25 February 2018 (2018-02-25), pages 93 - 98 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117115856B (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9576367B2 (en) | Object detection method and device | |
| US9805505B2 (en) | Estimation of object properties in 3D world | |
| CN108198044B (en) | Commodity information display method, commodity information display device, commodity information display medium and electronic equipment | |
| US8706663B2 (en) | Detection of people in real world videos and images | |
| US8995714B2 (en) | Information creation device for estimating object position and information creation method and program for estimating object position | |
| CN108027877A (en) | System and method for the detection of non-barrier | |
| CN105447529A (en) | Costume detection and attribute value identification method and system | |
| US20210122367A1 (en) | Methods, devices, media, and apparatuses of detecting moving object, and of intelligent driving control | |
| CN114758464B (en) | Storage battery anti-theft method, device and storage medium based on charging pile monitoring video | |
| CN111160220B (en) | Deep learning-based parcel detection method and device and storage medium | |
| KR20210099450A (en) | Far away small drone detection method Using Deep Learning | |
| Nghiem et al. | Background subtraction in people detection framework for RGB-D cameras | |
| JP2018120283A (en) | Information processing device, information processing method and program | |
| CN104508706B (en) | Feature extraction method, program and system | |
| JP2018503195A (en) | Object detection method and object detection apparatus | |
| CN117115856B (en) | Target detection method based on image fusion, human body security inspection equipment and storage medium | |
| JPWO2019215780A1 (en) | Identification system, model re-learning method and program | |
| JP2018197945A (en) | Obstacle detection apparatus and obstacle detection method | |
| JP6992900B2 (en) | Information processing equipment, control methods, and programs | |
| CN111260607A (en) | Automatic suspicious article detection method, terminal device, computer device and medium | |
| KR102407202B1 (en) | Apparatus and method for intelligently analyzing video | |
| KR20160039447A (en) | Spatial analysis system using stereo camera. | |
| JP2022019988A (en) | Information processing apparatus, display device, and control method | |
| KR102085034B1 (en) | Method and Apparatus for Detecting Foregroud Image with Separating Foregroud and Background in Image | |
| JP7349290B2 (en) | Object recognition device, object recognition method, and object recognition program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |