CN109241896B - Channel safety detection method and device and electronic equipment - Google Patents
Channel safety detection method and device and electronic equipment Download PDFInfo
- Publication number
- CN109241896B CN109241896B CN201810990426.4A CN201810990426A CN109241896B CN 109241896 B CN109241896 B CN 109241896B CN 201810990426 A CN201810990426 A CN 201810990426A CN 109241896 B CN109241896 B CN 109241896B
- Authority
- CN
- China
- Prior art keywords
- channel
- image
- target object
- detected
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 203
- 238000000034 method Methods 0.000 claims description 43
- 238000010801 machine learning Methods 0.000 claims description 12
- 238000004891 communication Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000000875 corresponding effect Effects 0.000 description 19
- 238000010586 diagram Methods 0.000 description 14
- 238000012545 processing Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 9
- 238000013527 convolutional neural network Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Alarm Systems (AREA)
- Closed-Circuit Television Systems (AREA)
- Emergency Alarm Devices (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a channel safety detection method, a device and electronic equipment, belonging to the technical field of intelligent security, wherein the channel safety detection method comprises the following steps: acquiring images of a channel to be detected, which are acquired according to a preset time interval; inputting the image into a target detection model for target detection to obtain target detection information of the image; and determining the safety detection result of the channel to be detected based on the target detection information of the image. The invention improves the reliability of the detection result while realizing the real-time safety detection of the channel to be detected, and can more effectively avoid the phenomenon that the channel is occupied or blocked.
Description
Technical Field
The invention relates to the technical field of intelligent security and protection, in particular to a channel security detection method and device and electronic equipment.
Background
At present, from the viewpoint of safety, special passages are often arranged, and the special passages are mainly used for escape or rescue in emergency situations, such as fire fighting passages, safe escape passages and the like.
However, at present, the phenomenon that the above-mentioned special channel is occupied or blocked occurs sometimes, for example, the fire fighting channel is used as a temporary parking space, a vehicle is parked on the fire fighting channel, and the fire fighting channel is occupied or blocked.
Therefore, an effective or reliable scheme is needed to realize real-time safety detection of the special channel and avoid the phenomenon that the channel is occupied or blocked.
Disclosure of Invention
In order to solve the problems in the prior art, embodiments of the present invention provide a method and an apparatus for detecting channel security, and an electronic device. The technical scheme is as follows:
in one aspect, a method for detecting channel security is provided, where the method includes:
acquiring images of a channel to be detected, which are acquired according to a preset time interval;
inputting the image into a target detection model for target detection to obtain target detection information of the image;
and determining the safety detection result of the channel to be detected based on the target detection information of the image.
In another aspect, an apparatus for detecting security of a channel is provided, the apparatus including:
the first acquisition module is used for acquiring images of a channel to be detected, which are acquired according to a preset time interval;
the target detection module is used for inputting the image into a target detection model for target detection to obtain target detection information of the image;
and the first detection result determining module is used for determining the safety detection result of the channel to be detected based on the target detection information of the image.
In another aspect, an electronic device is provided, including:
a processor adapted to implement one or more instructions; and the number of the first and second groups,
a memory storing one or more instructions adapted to be loaded by the processor and execute the channel security detection method described above.
In another aspect, a computer storage medium is provided, which stores computer program instructions, and when executed, the computer program instructions implement the channel security detection method described above.
The technical scheme provided by the embodiment of the invention has the following beneficial effects:
according to the method, the image of the channel to be detected, which is acquired according to the preset time interval, is acquired, the image is input into the target detection model for target detection, the target detection information of the image is acquired, and the safety detection result of the channel to be detected is determined based on the target detection information of the image.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flowchart of a channel security detection method according to an embodiment of the present invention;
fig. 2 is a schematic view of an application scenario of the channel security detection method according to the embodiment of the present invention;
FIG. 3 is a schematic diagram of an SSD model according to an embodiment of the present invention;
fig. 4 is a schematic flowchart of a process of determining a safety detection result of the channel to be detected based on the target detection information of the image according to the embodiment of the present invention;
FIG. 5 is a schematic flow chart of determining the associated object in the image according to the size information of the object according to the embodiment of the present invention;
fig. 6 is a schematic flow chart illustrating a process of determining whether the associated target is an unmoved target according to an embodiment of the present invention;
fig. 7 is a schematic flowchart of another channel security detection method according to an embodiment of the present invention;
fig. 8 is a schematic flowchart of determining a safety detection result of the channel to be detected based on the image and the background image according to the embodiment of the present invention;
fig. 9 is another schematic flow chart of determining a safety detection result of the channel to be detected based on the image and the background image according to the embodiment of the present invention;
fig. 10 is a schematic structural diagram of a channel safety detection device according to an embodiment of the present invention;
fig. 11 is a block diagram of a first detection result determining module according to an embodiment of the present invention;
fig. 12 is a block diagram of an associated object determining module according to an embodiment of the present invention;
fig. 13 is a block diagram of a first determining module according to an embodiment of the present invention;
fig. 14 is a schematic diagram of another apparatus for detecting security of a channel according to an embodiment of the present invention;
fig. 15 is a block diagram of a second detection result determining module according to an embodiment of the present invention;
fig. 16 is another block diagram of the second detection result determining module according to the embodiment of the present invention;
fig. 17 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
Please refer to fig. 1, which is a flowchart illustrating a channel security detection method according to an embodiment of the present invention. It should be noted that the channel security detection method according to the embodiment of the present invention may be applied to the channel security detection apparatus according to the embodiment of the present invention, and the channel security detection apparatus may be configured in an electronic device, where the electronic device may be a terminal or a server. The terminal can be a hardware device with various operating systems, such as a mobile phone, a tablet computer, a palm computer, a personal digital assistant and the like. The server may be one server or a server cluster composed of a plurality of servers.
Further, the present specification provides method steps as described in the examples or flowcharts, but may include more or fewer steps based on routine or non-inventive labor. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. In actual system or product execution, sequential execution or parallel execution (e.g., parallel processor or multi-threaded environment) may be possible according to the embodiments or methods shown in the figures. Specifically, as shown in fig. 1, the method may include:
s102, acquiring images of the channel to be detected, which are acquired according to a preset time interval.
In the present embodiment, the passage to be detected is a passage that is not allowed to be occupied or blocked, and may be, for example, a fire passage, a safety escape passage, or the like.
The preset time interval may be a length of time that allows the stagnation of the target object in the passage, and the target object may be determined according to a category of the passage to be detected, for example, when the passage to be detected is a fire passage, the target object may be determined as a vehicle. Accordingly, the preset time interval may be a length of time that the vehicle is allowed to stagnate in the fire passage; when the channel to be detected is a safe escape channel, the target object can be a moving object in the channel. Accordingly, the preset time interval may be a time period during which the moving object is allowed to stay in the safe escape route. In practical applications, the preset time interval may be set according to a corresponding safety rule, or may be set as needed on the premise that the safety rule is satisfied, for example, the preset time interval may be set to a time length shorter than the dead time length allowed by the safety rule, for example, the dead time length allowed by the safety rule is 5 minutes, and the preset time interval may be set to 2 seconds.
Specifically, as shown in fig. 2, which is an application scenario schematic diagram of the channel safety detection method provided in the embodiment of the present invention, an image acquisition device (for example, a camera) may be erected at a position of a channel to be detected, and the image acquisition device may be used to take a picture or take a video of the channel to be detected. When the image acquisition device is used for photographing the channel to be detected, photographing actions can be set according to the preset time interval, for example, the preset time interval is 10 minutes, the channel to be detected can be photographed once every 10 minutes, or can be photographed for multiple times within the preset time interval, and the image of the channel to be detected can be two images photographed at the preset time interval, or can be multiple images photographed within the preset time interval; (ii) a When the image acquisition device is used for video shooting of a channel to be detected, the image of the channel to be detected may be two images of corresponding frames acquired from the shot video according to a preset time interval, or may be an image of a plurality of frames within the preset time interval acquired from the shot video, for example, the preset time interval is 10 minutes, then two images of the channel to be detected with the time interval of 10 minutes may be acquired from the shot video, or a plurality of images of any frame between two frames with the time interval of 10 minutes may be acquired from the shot video.
And S104, inputting the image into a target detection model for target detection to obtain target detection information of the image.
In the embodiments of the present specification, the object detection model may directly detect an object in the input image to output object detection information corresponding to the input image, and the object detection information may include category information and position information.
The target detection model is obtained by training a preset machine learning model by adopting a deep learning method. The preset machine learning model can be a convolutional neural network R-CNN or Fast R-CNN, and can also be a convolutional neural network YOLO or SSD. The convolutional neural network YOLO or SSD can express the detection task as an end-to-end regression problem, and corresponding position information and category information can be obtained simultaneously by processing the image once, so that the processing speed is high, the generalization performance is good, and the real-time processing can be realized. Therefore, it is preferable in the embodiment of the present specification to use the convolutional neural network YOLO or SSD as the preset machine learning model.
In a specific embodiment, the preset machine learning model may be an SSD model, the SSD model is a deep learning model applied to target detection, and as shown in fig. 3, the structural schematic diagram of the SSD model provided in the embodiment of the present invention is improved based on the VGG16 image classification model structure, two full connection layers of the VGG16 are changed into convolutional layers to obtain a basic convolutional network of the SSD, and then a multi-scale feature mapping layer composed of 4 convolutional layers is added after the basic convolutional network, the size of the added 4 convolutional layers is gradually reduced to ensure multi-scale of the predicted detection result, and each added convolutional layer may generate a series of detected detection results through a series of filters. The base convolutional network and the multi-scale feature mapping layer form an SSD model.
After the preset machine learning model is determined, the preset machine learning model needs to be trained based on the marked sample image containing the target object, and in the training process, the model parameters of the preset machine learning model are adjusted until the result output by the preset machine learning model converges, so that the machine learning model corresponding to the model parameters in the converging process can be determined as the target detection model.
For the marked sample image, the marked image containing the target object can be directly used in some open source databases (such as MySQL database, PostgreSQL database, and the like), or the marked image containing the target object can be marked by a marking tool after the image actually containing the target object is shot by itself. For example, the position information of the target object in the sample image may be marked by a multi-dimensional vector, such as the position information of the target object in the sample image may be marked by a four-dimensional vector (x, y, w, h). Wherein x may represent a horizontal axis coordinate of the center point of the object, y may represent a vertical axis coordinate of the center point of the object, w may represent a width of the object, and h may represent a height of the object.
It should be noted that, in order to improve the stability of the trained target detection model and ensure the reliability of the detection result, image enhancement processing may be performed on the sample image to avoid the influence of outdoor light, such as adjusting the brightness and saturation of the sample image and performing noise addition processing, and then the sample image after the image enhancement processing is input into a preset machine learning model to perform model training.
In practical application, a channel to be detected is often located in a certain area in an acquired image, and other areas of the image often contain a lot of other image information.
The preset channel region may be obtained by framing the acquired image with a response processing function and a callback function provided by an OpenCV (open source computer vision library), or by identifying the channel to be detected in the image by using an automatic identification method.
Specifically, after an image is input into the target detection model, feature extraction may be performed through a basic convolution network of the target detection model to obtain outputs of each convolution layer, and then, the outputs of 5 different convolution layers are convolved with two added convolution layers, respectively, one convolution outputs a confidence for classification, and the other convolution outputs a position coordinate (x, y, w, h) for regression, where x may represent a horizontal axis coordinate of a center point of the target object, y may represent a vertical axis coordinate of the center point of the target object, w may represent a width of the target object, and h may represent a height of the target object. And finally, integrating the prediction results and obtaining target detection information corresponding to the input image through a non-maximum suppression algorithm.
S106, determining the safety detection result of the channel to be detected based on the target detection information of the image.
Specifically, the target detection information may include position information and size information of a target object in the image. The position information of the target object may be an abscissa and an ordinate (x, y) of a central point of the target object in the image, and the size information of the target object may be a width w and a height h of the target object in the image, that is, (w, h).
Correspondingly, as shown in fig. 4, a schematic flow chart for determining a safety detection result of the channel to be detected based on the target detection information of the image according to the embodiment of the present invention is provided. As shown in fig. 4, the method may include:
s402, determining the related target object in the image according to the size information of the target object.
In the embodiment of the present specification, since each image input to the target detection model may include a plurality of target objects, in order to correspond the same target object in each image, it is necessary to find out the associated target object in each image according to the size information of each target object.
Specifically, a method for determining the associated object in the image according to the size information of the object may be as shown in fig. 5. Fig. 5 is a schematic flowchart illustrating a process of determining an associated object in the image according to size information of the object according to an embodiment of the present invention, as shown in fig. 5, the method may include:
s502, calculating the difference size of the target object according to the size information of the target object in the image.
Specifically, the width difference and the height difference between each target object in the image after the time point and the target object in the image before the time point are calculated, and the difference size corresponding to the target object is obtained. The image with the prior time point is the image of the channel to be detected, which is acquired before the image with the later time point.
In the embodiments of the present specification, the difference size of the target object can be calculated by using the following formula:
wherein, i is a target object in the image with the previous time point in the image of the channel to be detected; j is a target object in the image with the later time point in the image of the channel to be detected;
w i the width value of the target object i in the image before the time point; w is a j The width value of the target object j in the image after the time point is obtained; Δ w i-j Is the difference in width between object i and object j.
h i Is the height value of the object i in the image before the time point; h is j The height value of the target object j in the image after the time point; Δ h i-j Is the height difference between object i and object j.
It should be noted that the above is only an example of calculating the difference size of the target object in the image of the channel to be detected, and in practical applications, the difference size may also be calculated in other manners, for example, a value may be further calculated according to the width difference and the height difference, and the value is taken as the difference size.
S504, judging whether the difference size of the target object is smaller than a first preset threshold value.
In the embodiment of the present specification, the form of the first preset threshold may be set according to the form of the difference size of the target object, for example, if the difference size of the target object includes a width difference value and a height difference value, the first preset threshold may be correspondingly set to include a width difference threshold and a height difference threshold; if the difference size of the target object is a value calculated according to the width difference and the height difference, the first threshold may be correspondingly set as a value threshold.
In addition, the size of the first preset threshold may be set in combination with the detection accuracy of the target detection model in practical applications. In general, the higher the detection accuracy of the object detection model is, the smaller the first preset threshold value can be set.
Specifically, if the difference size of the objects is (Δ w) i-j ,Δh i-j ) The first preset threshold is (Δ w) s ,Δh s ) In this step, Δ w needs to be determined simultaneously i-j Whether or not less than Δ w s And Δ h i-j Whether or not less than Δ h s . When Δ w i-j <Δw s And Δ h i-j <Δh s Then, it may be determined that the difference size between the object i and the object j is smaller than the first preset threshold, and step S306 is performed.
S506, when the judgment result is yes, determining the target object corresponding to the difference size as the related target object.
Specifically, when the judgment result is that the difference size of the target object is smaller than the first preset threshold, the target object corresponding to the difference size may be determined to be the associated target object. For example, when Δ w i-j <Δw s And Δ h i-j <Δh s Then, it may be determined that the object i in the image before the time point and the object j in the image after the time point are associated objects. Otherwise, when the judgment result is that the difference size of the target object is greater than or equal to the first preset threshold, the target corresponding to the difference size can be determinedThe object is a non-associated object.
S404, judging whether the associated target object is an unmoved target object according to the position information of the associated target object in each image.
In the embodiments of the present specification, the position information of the object in the image may be expressed as coordinate values (x, y) of the center point of the object in the image.
Specifically, a method for determining whether the associated target object is an unmoved target object according to the position information of the associated target object in each image may be as shown in fig. 6. Fig. 6 is a schematic flowchart illustrating a process of determining whether the associated target object is an unmoved target object according to an embodiment of the present invention, as shown in fig. 6, the method may include:
s602, calculating the offset value of the related target object according to the position information of the related target object in each image.
In this embodiment, the offset value of the associated target object may be calculated by the following formula:
wherein,
Δz i-j an offset value representing the associated object i and object j;
(x i ,y i ) Coordinate values representing the center point of the target object i in the image;
(x j ,y j ) Coordinate values representing the center point of the target object j in the image; and the target object i and the target object j are related target objects in the image of the channel to be detected.
It should be noted that the above is only an example of calculating the offset value of the associated target object in the image, and the offset value may be calculated in other manners according to practical applications, and the present invention is not limited thereto.
S604, judging whether the deviation value of the associated target object is smaller than a second preset threshold value.
Specifically, the second preset threshold may be set to t i Multiple of, i.e. the second predetermined threshold value deltaz s Can be set to Δ z s =ρ·t i . The coefficient ρ can be set according to the length of a preset time interval in practical application, that is, the time interval of image acquisition of the channel to be detected. Generally, the shorter the preset time interval is, the smaller the coefficient ρ is; conversely, the longer the preset time interval, the larger the coefficient ρ. For example, when the preset time interval is set to 1 second, the coefficient ρ may be set to 0.5; when the preset time interval is set to 5 seconds, the coefficient ρ may be set to 2.
Calculating the offset value delta z of the associated target object in the image of the channel to be detected i-j Then, the offset value Δ z of the target object to be correlated is calculated i-j And a second predetermined threshold value deltaz s Comparing when the value is delta z i-j <Δz s Then, it may be determined that the offset value of the associated target object is less than the second preset threshold value, and step S406 may be performed. When Δ z is i-j ≥Δz s It may be determined that the offset value of the associated target object is greater than or equal to a second preset threshold.
S606, when the judgment result is yes, the associated target object is determined to be the unmoved target object.
Specifically, when the determination result is that the offset value of the associated target object is smaller than the second preset threshold, it may be determined that the associated target object is an unmoved target object. For example, when Δ z i-j <Δz s Then, the associated object i and object j may be determined to be unmoved objects. Otherwise, when the judgment result is that the deviation value of the associated target object is greater than or equal to the second preset threshold, it indicates that the associated target object is moving.
S406, when the judgment result is yes, determining that the safety detection result of the channel to be detected is a non-safety channel.
Specifically, if the result of the determination is that the associated target object is an unmoved target object, it indicates that the target object does not move in the channel to be detected within the preset time interval, that is, the target object is parked on the channel to be detected, so that the channel to be detected is occupied or blocked, and it can be determined that the safety detection result of the channel to be detected is an unsafe channel.
If the judgment result shows that the associated target object is the moving target object, the judgment result shows that the target object passes through the channel to be detected within the preset time interval, and the target object is not parked on the channel to be detected, so that the target object cannot occupy or block the channel to be detected. When all the related target objects are moving target objects, it is indicated that no unmoving target object exists on the channel to be detected, and at this time, it can be determined that the channel to be detected is currently a safe channel.
In the embodiment of the description, when it is determined that the safety detection result of the channel to be detected is a non-safety channel, alarm information can be output to inform a manager to remove an unmoved target object from the channel to be detected in time, so that the smoothness of the channel to be detected is ensured.
To sum up, in the embodiment of the present invention, the image of the channel to be detected acquired according to the preset time interval is acquired, the image is input into the target detection model for target detection, the target detection information of the image is acquired, and the safety detection result of the channel to be detected is determined based on the target detection information of the image.
In addition, the method of the embodiment of the invention only needs to detect the currently acquired image without combining a background image set, and the background image set usually contains a large number of background images, so that the data processing amount of the method is much smaller, and the method has higher real-time performance and higher processing speed due to the high processing speed of the target detection model, and can more effectively avoid the phenomenon that the channel is occupied or blocked.
Fig. 7 is a schematic flow chart illustrating another channel security detection method according to an embodiment of the invention. The present specification provides method steps as described in the examples or flowcharts, but may include more or fewer steps based on routine or non-inventive practice. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. In actual system or product execution, sequential execution or parallel execution (e.g., parallel processor or multi-threaded environment) may be used according to the embodiments or methods shown in the figures. Specifically, as shown in fig. 7, the method may include:
s702, acquiring images of the channel to be detected, which are acquired according to a preset time interval.
Specifically, the step may refer to the related description in the embodiment of the method shown in fig. 1, and is not described herein again.
S704, inputting the image into a target detection model for target detection to obtain target detection information of the image.
Specifically, the step may refer to the related description in the embodiment of the method shown in fig. 1, and is not described herein again.
It should be noted that, since the target detection model can only detect the target object in the image, when no target object exists in the image, the target detection information output by the target detection model will not include any information of the target object.
S706, judging whether the image contains the target object.
Specifically, whether the image includes the target object may be determined based on target detection information output by the target detection model. When the output target detection information does not contain any information of the target object, determining that no target object exists in the image of the channel to be detected; conversely, when the output target detection information includes information of the target object, it can be determined that the target object exists in the image of the channel to be detected.
When it is determined that no target object exists in the image of the channel to be detected, steps S708 to S710 may be performed. Step S712 may be performed when it is determined that the target object exists in the image of the channel to be detected.
S708, when the target object is not contained in the image, obtaining a background image of the channel to be detected.
In the embodiment of the present specification, the background image is an image of the channel to be detected in a clear state, and the clear state is a state of the channel to be detected without any obstacle. The background image of the channel to be detected can be collected and stored in advance through a set image collecting device.
S710, determining a safety detection result of the channel to be detected based on the image and the background image of the channel to be detected.
Considering that other objects may exist on the channel to be detected without the target object, and the other objects may also occupy or block the channel to be detected, when it is determined that the image of the channel to be detected does not contain the target object, the image of the channel to be detected and the background image may be further combined to determine the safety detection result of the channel to be detected.
In some embodiments, as shown in fig. 8, fig. 8 is a schematic flow chart of determining a safety detection result of the channel to be detected based on the image of the channel to be detected and the background image according to an embodiment of the present invention, which specifically includes:
s802, calculating a pixel difference value between a first pixel point in the image of the channel to be detected and a second pixel point in the background image.
Specifically, the first pixel point is any pixel point in the image of the channel to be detected, the second pixel point is any pixel point in the background image, and the pixel difference value between the first pixel point in the image of the channel to be detected and the second pixel point in the background image is calculated respectively.
S804, determining a connected region according to the pixel points with the pixel difference value larger than the preset pixel difference value.
Specifically, the preset pixel difference value may be set according to the reliability requirement of the detection result, and generally, the higher the reliability requirement of the detection result is, the larger the preset pixel difference value may be set; conversely, the lower the reliability requirement of the detection result is, the smaller the preset pixel difference value can be set.
The corresponding foreground image can be obtained through the pixel difference calculation of the image of the channel to be detected and the background image, and the adjacent pixel points in the foreground image with the pixel difference larger than the preset pixel difference form a communicated area. The area of each connected region can be further calculated through a connected region algorithm.
S806, judging whether the area of the connected region is larger than a preset area.
In the embodiments of the present specification, the preset area may be set according to the maximum area determined as the non-obstacle object. When the area of the communication area is larger than the preset area, the object in the channel to be detected corresponding to the communication area is considered as an obstacle, and the channel to be detected may be occupied or blocked; when the area of the communication area is smaller than or equal to the preset area, the object in the channel to be detected corresponding to the communication area is not considered to be an obstacle, and the channel to be detected cannot be occupied or blocked.
And S808, when the area of the communication area is larger than the preset area, determining that the safety detection result of the channel to be detected is a non-safety channel.
Specifically, when the area of the communication region is larger than the preset area, it is indicated that the object in the to-be-detected channel corresponding to the communication region is an obstacle and may occupy or block the to-be-detected channel, and at this time, it may be determined that the safety detection result of the to-be-detected channel is a non-safety channel. Correspondingly, when the obstacle is determined to be a non-safe channel, alarm information can be output, so that a manager can remove the obstacle from the channel to be detected in time.
When the area of the non-existence communicating area is larger than the preset area, it is indicated that no barrier exists in the channel to be detected, and at the moment, the safety detection result of the channel to be detected can be determined to be a safety channel.
In other embodiments, as shown in fig. 9, fig. 9 is another schematic flow chart of determining a safety detection result of the channel to be detected based on the image of the channel to be detected and the background image according to an embodiment of the present invention, which specifically includes:
and S902, calculating the matching degree of the image of the channel to be detected and the background image.
Specifically, the matching degree between each image in the image of the channel to be detected and the background image may be calculated, and in the embodiment of the present specification, the matching degree between two images may include a numerical value obtained by quantizing the similarity between the images according to a certain rule.
In a specific embodiment, the matching degree between the two images may include a pixel matching degree, that is, a quantized value of a degree of similarity between image pixels, and correspondingly, the calculating the matching degree between the image of the channel to be detected and the background image may include: and respectively calculating the pixel matching degree between each image in the image of the channel to be detected and the background image.
In another specific embodiment, the degree of matching between two images may include a feature point matching degree, i.e., a quantified value of the degree of similarity between feature points of the images. Specifically, when a quantization value of the degree of similarity between image feature points is calculated, feature vectors of the image feature points may be extracted, and then the matching degree between two images may be determined by calculating the distance between the feature vectors. Specifically, the distance may include, but is not limited to, a euclidean distance, a cosine distance, a manhattan distance, and the like. Correspondingly, the calculating the matching degree between the image of the channel to be detected and the background image may include: and respectively calculating the matching degree of the characteristic points between each image and the background image in the image of the channel to be detected.
In addition, it should be noted that, in the embodiment of the present invention, the matching degree between the two images may include, but is not limited to, a pixel matching degree and/or a feature point matching degree, and in practical applications, other quantitative values capable of characterizing the similarity degree between the two images may also be included.
And S904, judging whether the matching degree is smaller than a preset matching degree.
In the embodiment of the present specification, the preset matching degree may be set according to the reliability requirement for the detection result, and generally, the higher the reliability requirement for the detection result is, the larger the preset matching degree may be set; conversely, the lower the reliability requirement for the detection result, the smaller the preset matching degree can be set.
Specifically, the matching degrees corresponding to the images of the channel to be detected can be respectively compared with the preset matching degrees to judge whether the corresponding matching degrees are all smaller than the preset matching degrees, and when all the matching degrees corresponding to the images of the channel to be detected are smaller than the preset matching degrees, the matching degrees can be determined to be smaller than the preset matching degrees; of course, the maximum matching degree may be selected from all matching degrees corresponding to the image of the channel to be detected, and then the maximum matching degree is compared with the preset matching degree, and when the maximum matching degree is smaller than the preset matching degree, it may be determined that the matching degree is smaller than the preset matching degree.
S906, when the matching degree is smaller than a preset matching degree, determining that the channel to be detected is a non-safe channel.
Specifically, when the matching degree is smaller than the preset matching degree as a result of the judgment, it is indicated that the image of the channel to be detected and the background image are different, an obstacle exists in the image of the channel to be detected, it can be determined that the safety detection result of the channel to be detected is a non-safety channel, and correspondingly, alarm information can be output when the non-safety channel is determined, so that a manager can timely remove the obstacle from the channel to be detected. On the contrary, when the matching degree is larger than or equal to the preset matching degree as a result of the judgment, it is indicated that the image of the channel to be detected is similar to the background image, no obstacle exists in the image of the channel to be detected, and the detection result of the channel to be detected at the moment can be determined to be a safe channel.
In addition, it should be noted that the above are only two examples of determining the safety detection result of the channel to be detected based on the image of the channel to be detected and the background image, and in practical applications, other methods may be included in combination with practical application requirements.
S712, determining the safety detection result of the channel to be detected based on the target detection information of the image of the channel to be detected.
Specifically, the step may refer to the related description in the embodiment of the method shown in fig. 1, and is not described herein again.
To sum up, the embodiment of the invention acquires the image of the channel to be detected acquired according to the preset time interval, inputs the image into the target detection model for target detection, acquires the target detection information of the image, determines the safety detection result based on the target detection information when the target detection information shows that the target object exists in the image of the channel to be detected, and determines the safety detection result by combining the background image of the channel to be detected when the target detection information shows that the target object does not exist in the image of the channel to be detected, thereby greatly improving the reliability of the detection result on the basis of realizing real-time and quick detection, and more effectively avoiding the phenomenon that the channel is occupied or blocked.
The channel security detection apparatus provided in the embodiment of the present invention corresponds to the channel security detection methods provided in the embodiments described above, so that the implementation of the channel security detection method described above is also applicable to the channel security detection apparatus provided in this embodiment, and a detailed description is not given in this embodiment.
Referring to fig. 10, a schematic structural diagram of a channel security detection apparatus according to an embodiment of the present invention is shown, and as shown in fig. 10, the apparatus may include:
the first obtaining module 1010 may be configured to obtain images of a channel to be detected, which are collected according to a preset time interval;
the target detection module 1020 may be configured to input the image into a target detection model for target detection, so as to obtain target detection information of the image;
a first detection result determining module 1030, configured to determine a safety detection result of the channel to be detected based on the target detection information of the image.
In a specific embodiment, the target detection information includes position information and size information of a target object in the image; accordingly, as shown in fig. 11, the first detection result determining module 1030 may include:
a related object determining module 1031, configured to determine a related object in the image according to the size information of the object;
the first determining module 1032 may be configured to determine whether the associated target object is an unmoved target object according to the position information of the associated target object in each image;
the first determining module 1033 may be configured to determine that a security detection result of the channel to be detected is a non-security channel when the associated target object is an unmoved target object.
Optionally, as shown in fig. 12, the association target object determining module 1031 may include:
a first calculating module 1201, which may be configured to calculate a difference size of an object in the image according to size information of the object;
a second determining module 1202, configured to determine whether the difference size of the target object is smaller than a first preset threshold;
the second determining module 1203 may be configured to determine, when the difference size of the target object is smaller than a first preset threshold, that the target object corresponding to the difference size is a related target object.
Optionally, as shown in fig. 13, the first determining module 1032 may include:
a second calculating module 1301, which may be configured to calculate an offset value of the associated target object according to the position information of the associated target object in each image;
a third determining module 1302, configured to determine whether the offset value of the associated target object is smaller than a second preset threshold;
the third determining module 1303 may be configured to determine that the associated target object is an unmoved target object when the offset value of the associated target object is smaller than a second preset threshold.
Referring to fig. 14, a schematic structural diagram of another channel safety detection device provided in the present invention is shown, and as shown in fig. 14, the device may include:
the first obtaining module 1410 may be configured to obtain an image of a channel to be detected, where the image is collected according to a preset time interval;
the target detection module 1420 may be configured to input the image into a target detection model to perform target detection, so as to obtain target detection information of the image;
a first detection result determining module 1430, configured to determine a safety detection result of the channel to be detected based on the target detection information of the image;
the fourth determining module 1440 may be configured to determine whether the image includes the target object before determining the safety detection result of the channel to be detected based on the target detection information of the image.
In one embodiment, as shown in fig. 14, the apparatus may further include:
a second obtaining module 1450, configured to obtain a background image of the channel to be detected when the image does not include the target object; the background image is an image when the channel to be detected is in a smooth state;
the second detection result determining module 1460 may be configured to determine a safety detection result of the channel to be detected based on the image and the background image.
Optionally, as shown in fig. 15, the second detection result determining module 1460 may include:
a third calculating module 1461, configured to calculate a pixel difference between a first pixel point in the image and a second pixel point in the background image;
the region determining module 1462 may be configured to determine a connected region according to a pixel point of which the pixel difference value is greater than a preset pixel difference value;
a fifth determining module 1463, configured to determine whether an area of the connected region is larger than a preset area;
the fourth determining module 1464 may be configured to determine that the safety detection result of the channel to be detected is a non-safety channel when the area of the communication region is greater than the preset area.
Optionally, as shown in fig. 16, the second detection result determining module 1460 may include:
a fourth calculating module 1465, configured to calculate a matching degree between the image and the background image;
a sixth determining module 1466, configured to determine whether the matching degree is smaller than a preset matching degree;
a fifth determining module 1467, configured to determine that the channel to be detected is a non-secure channel when the matching degree is smaller than a preset matching degree.
It should be noted that, when the apparatus provided in the foregoing embodiment implements the functions thereof, only the division of the functional modules is illustrated, and in practical applications, the functions may be distributed by different functional modules according to needs, that is, the internal structure of the apparatus may be divided into different functional modules to implement all or part of the functions described above.
Fig. 17 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, where the electronic device is used to implement the channel security detection method provided in the foregoing embodiment. The electronic device may be a terminal device such as a PC (personal computer), a mobile phone, a PDA (tablet personal computer), or a service device such as an application server and a cluster server. Referring to fig. 15, the internal structure of the electronic device may include, but is not limited to: a processor, a network interface, and a memory. The processor, the network interface, and the memory in the electronic device may be connected by a bus or in other manners, and fig. 15 shown in the embodiment of the present specification is exemplified by being connected by a bus.
The processor (or CPU) is a computing core and a control core of the electronic device. The network interface may optionally include a standard wired interface, a wireless interface (e.g., WI-FI, mobile communication interface, etc.). A Memory (Memory) is a Memory device in an electronic device for storing programs and data. It is understood that the memory herein may be a high-speed RAM storage device, or may be a non-volatile storage device (non-volatile memory), such as at least one magnetic disk storage device; optionally, at least one memory device located remotely from the processor. The memory provides a storage space that stores an operating system of the electronic device, which may include, but is not limited to: a Windows system (an operating system), a Linux system (an operating system), an Android system, an IOS system, etc., which are not limited in the present invention; also, one or more instructions, which may be one or more computer programs (including program code), are stored in the memory space and are adapted to be loaded and executed by the processor. In this embodiment of the present specification, the processor loads and executes one or more instructions stored in the memory to implement the channel security detection method provided in the foregoing method embodiment.
Embodiments of the present invention also provide a storage medium, which may be disposed in an electronic device to store at least one instruction, at least one program, a code set, or a set of instructions related to implementing a channel security detection method in the method embodiments, where the at least one instruction, the at least one program, the code set, or the set of instructions may be loaded and executed by a processor of the electronic device to implement the channel security detection method provided in the method embodiments.
Optionally, in this embodiment, the storage medium may include but is not limited to: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (12)
1. A method for detecting channel security, the method comprising:
acquiring images of a channel to be detected, which are acquired according to a preset time interval;
inputting the image into a target detection model for target detection to obtain target detection information of the image; the target detection information comprises position information and size information of a target object in the image, and the size information comprises a width value and a height value; the target detection model is obtained by training a preset machine learning model based on a marked sample image containing a target object;
determining a safety detection result of the channel to be detected based on the target detection information of the image; the safety detection result indicates that the channel to be detected is a safety channel or a non-safety channel;
the determining the safety detection result of the channel to be detected based on the target detection information of the image comprises:
respectively calculating the width difference value and the height difference value of the target object according to the width value and the height value of the target object in the image after the time point and the target object in the image before the time point; when the width difference value is smaller than a width difference threshold value and the height difference value is smaller than a height difference threshold value, determining that the target object corresponding to the width difference value and the height difference value is a related target object;
judging whether the associated target object is an unmoved target object or not according to the position information of the associated target object in each image;
and when the judgment result is yes, determining that the safety detection result of the channel to be detected is a non-safety channel.
2. The channel safety detection method according to claim 1, wherein the determining whether the associated object is an unmoved object according to the position information of the associated object in each image comprises:
calculating an offset value of the associated target object according to the position information of the associated target object in each image;
judging whether the deviation value of the associated target object is smaller than a second preset threshold value or not;
and when the judgment result is yes, determining that the associated target object is an unmoved target object.
3. The channel security detection method according to claim 1, wherein before determining the security detection result of the channel to be detected based on the target detection information of the image, the method further comprises:
judging whether the image contains a target object;
and when the image contains the target object, executing a step of determining a safety detection result of the channel to be detected based on the target detection information of the image.
4. The channel security detection method of claim 3, further comprising:
when the image does not contain the target object, acquiring a background image of the channel to be detected; the background image is an image of the channel to be detected in a smooth state;
and determining the safety detection result of the channel to be detected based on the image and the background image.
5. The channel security detection method according to claim 4, wherein the determining the security detection result of the channel to be detected based on the image and the background image comprises:
calculating a pixel difference value between a first pixel point in the image and a second pixel point in the background image;
determining a connected region according to the pixel point of which the pixel difference value is greater than a preset pixel difference value;
judging whether the area of the communication area is larger than a preset area or not;
and when the area of the communication area is larger than the preset area, determining that the safety detection result of the channel to be detected is a non-safety channel.
6. The channel security detection method according to claim 4, wherein the determining the security detection result of the channel to be detected based on the image and the background image comprises:
calculating the matching degree of the image and the background image;
judging whether the matching degree is smaller than a preset matching degree;
and when the matching degree is smaller than the preset matching degree, determining that the safety detection result of the channel to be detected is a non-safety channel.
7. A passageway safety detection device, the device comprising:
the first acquisition module is used for acquiring images of a channel to be detected, which are acquired according to a preset time interval;
the target detection module is used for inputting the image into a target detection model for target detection to obtain target detection information of the image; the target detection information comprises position information and size information of a target object in the image, and the size information comprises a width value and a height value; the target detection model is obtained by training a preset machine learning model based on a marked sample image containing a target object;
the first detection result determining module is used for determining the safety detection result of the channel to be detected based on the target detection information of the image; the safety detection result indicates that the channel to be detected is a safety channel or a non-safety channel;
the first detection result determination module includes:
the related target object determining module is used for respectively calculating the width difference value and the height difference value of the target object according to the width value and the height value of the target object in the image after the time point and the target object in the image before the time point; when the width difference value is smaller than a width difference threshold value and the height difference value is smaller than a height difference threshold value, determining the target object corresponding to the width difference value and the height difference value as a related target object;
the first judgment module is used for judging whether the associated target object is an unmoved target object according to the position information of the associated target object in each image;
and the first determining module is used for determining that the safety detection result of the channel to be detected is a non-safety channel when the associated target object is an unmoved target object.
8. The apparatus according to claim 7, wherein the first determining module comprises:
the second calculation module is used for calculating the offset value of the associated target object according to the position information of the associated target object in each image;
the third judging module is used for judging whether the deviation value of the associated target object is smaller than a second preset threshold value or not;
and the third determining module is used for determining that the associated target object is an unmoved target object when the deviation value of the associated target object is smaller than a second preset threshold value.
9. The apparatus for channel security detection as claimed in claim 7, further comprising:
and the fourth judging module is used for judging whether the image contains the target object or not before determining the safety detection result of the channel to be detected based on the target detection information of the image.
10. The apparatus for channel security detection as claimed in claim 9, further comprising:
the second acquisition module is used for acquiring a background image of the channel to be detected when the image does not contain the target object; the background image is an image when the channel to be detected is in a smooth state;
and the second detection result determining module is used for determining the safety detection result of the channel to be detected based on the image and the background image.
11. An electronic device, comprising:
a processor adapted to implement one or more instructions; and the number of the first and second groups,
memory storing one or more instructions adapted to be loaded by the processor and to perform the method of channel security detection according to any of claims 1-6.
12. A computer storage medium storing computer program instructions which, when executed, implement the channel security detection method of any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810990426.4A CN109241896B (en) | 2018-08-28 | 2018-08-28 | Channel safety detection method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810990426.4A CN109241896B (en) | 2018-08-28 | 2018-08-28 | Channel safety detection method and device and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109241896A CN109241896A (en) | 2019-01-18 |
CN109241896B true CN109241896B (en) | 2022-08-23 |
Family
ID=65068584
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810990426.4A Active CN109241896B (en) | 2018-08-28 | 2018-08-28 | Channel safety detection method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109241896B (en) |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110910355A (en) * | 2019-11-07 | 2020-03-24 | 浙江大华技术股份有限公司 | Package blocking detection method and device and computer storage medium |
CN111259763B (en) * | 2020-01-13 | 2024-02-02 | 华雁智能科技(集团)股份有限公司 | Target detection method, target detection device, electronic equipment and readable storage medium |
CN111311923A (en) * | 2020-03-05 | 2020-06-19 | 浙江长元科技有限公司 | Fire fighting access supervision method and system |
CN111582056A (en) * | 2020-04-17 | 2020-08-25 | 上善智城(苏州)信息科技有限公司 | Automatic detection method for fire-fighting equipment in oil unloading operation site of gas station |
CN111614938B (en) * | 2020-05-14 | 2021-11-02 | 杭州海康威视系统技术有限公司 | Risk identification method and device |
CN111626204B (en) * | 2020-05-27 | 2022-01-11 | 汪海洋 | Railway foreign matter invasion monitoring method and system |
CN111681424B (en) * | 2020-06-06 | 2022-03-04 | 浙江长元科技有限公司 | Method for processing illegal parking occupation event of fire fighting truck channel |
CN111723767B (en) * | 2020-06-29 | 2023-08-08 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and computer storage medium |
CN112052821B (en) * | 2020-09-15 | 2023-07-07 | 浙江智慧视频安防创新中心有限公司 | Fire-fighting channel safety detection method, device, equipment and storage medium |
CN112396040B (en) * | 2021-01-19 | 2022-03-01 | 成都四方伟业软件股份有限公司 | Method and device for identifying lane occupation of vehicle |
CN113033367A (en) * | 2021-03-18 | 2021-06-25 | 山东渤聚通云计算有限公司 | Monitoring method and device based on fire-fighting site and electronic equipment |
CN113627412A (en) * | 2021-08-02 | 2021-11-09 | 北京百度网讯科技有限公司 | Target area detection method, target area detection device, electronic equipment and medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107134144A (en) * | 2017-04-27 | 2017-09-05 | 武汉理工大学 | A kind of vehicle checking method for traffic monitoring |
CN107705552A (en) * | 2016-08-08 | 2018-02-16 | 杭州海康威视数字技术股份有限公司 | A kind of Emergency Vehicle Lane takes behavioral value method, apparatus and system |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102779348B (en) * | 2012-06-20 | 2015-01-07 | 中国农业大学 | Method for tracking and measuring moving targets without marks |
US9621760B2 (en) * | 2013-06-07 | 2017-04-11 | Digimarc Corporation | Information coding and decoding in spectral differences |
CN104240222A (en) * | 2013-06-19 | 2014-12-24 | 贺江涛 | Intelligent detecting method and device for firefighting access blockage |
CN103366374B (en) * | 2013-07-12 | 2016-04-20 | 重庆大学 | Based on the passageway for fire apparatus obstacle detection method of images match |
CN103364034B (en) * | 2013-08-08 | 2016-02-03 | 保定市天河电子技术有限公司 | A kind of track collapse rock-fall detection method and system |
CN103473792A (en) * | 2013-09-11 | 2013-12-25 | 清华大学深圳研究生院 | Method for detecting moving target |
CN105894702B (en) * | 2016-06-21 | 2018-01-16 | 南京工业大学 | Intrusion detection alarm system based on multi-camera data fusion and detection method thereof |
CN107728633B (en) * | 2017-10-23 | 2020-12-18 | 广州极飞科技有限公司 | Method and device for acquiring position information of target object, mobile device and control method thereof |
CN108171938A (en) * | 2017-12-08 | 2018-06-15 | 广州粤华物业有限公司 | Channel blockage monitoring and alarming system |
CN108320510B (en) * | 2018-04-03 | 2020-12-04 | 深圳市智绘科技有限公司 | Traffic information statistical method and system based on aerial video shot by unmanned aerial vehicle |
-
2018
- 2018-08-28 CN CN201810990426.4A patent/CN109241896B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107705552A (en) * | 2016-08-08 | 2018-02-16 | 杭州海康威视数字技术股份有限公司 | A kind of Emergency Vehicle Lane takes behavioral value method, apparatus and system |
CN107134144A (en) * | 2017-04-27 | 2017-09-05 | 武汉理工大学 | A kind of vehicle checking method for traffic monitoring |
Also Published As
Publication number | Publication date |
---|---|
CN109241896A (en) | 2019-01-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109241896B (en) | Channel safety detection method and device and electronic equipment | |
KR102122348B1 (en) | Method and device for face in-vivo detection | |
TWI677825B (en) | Method of video object tracking and apparatus thereof and non-volatile computer readable storage medium | |
CN111178183B (en) | Face detection method and related device | |
WO2021051601A1 (en) | Method and system for selecting detection box using mask r-cnn, and electronic device and storage medium | |
KR101891887B1 (en) | Image processing method, apparatus for predicting dangerous situation and method, server for predicting dangerous situation using thereof | |
JP6547744B2 (en) | Image processing system, image processing method and program | |
KR101963404B1 (en) | Two-step optimized deep learning method, computer-readable medium having a program recorded therein for executing the same and deep learning system | |
CN111553302B (en) | Key frame selection method, device, equipment and computer readable storage medium | |
US10909388B2 (en) | Population density determination from multi-camera sourced imagery | |
CN112528974B (en) | Distance measuring method and device, electronic equipment and readable storage medium | |
CN109447022B (en) | Lens type identification method and device | |
CN111814776B (en) | Image processing method, device, server and storage medium | |
KR102333520B1 (en) | Method, device and system for detecting object on road | |
CN113012383A (en) | Fire detection alarm method, related system, related equipment and storage medium | |
CN111507320A (en) | Detection method, device, equipment and storage medium for kitchen violation behaviors | |
CN110795599B (en) | Video emergency monitoring method and system based on multi-scale graph | |
CN111783910A (en) | Building project management method, electronic equipment and related products | |
CN113808200B (en) | Method and device for detecting moving speed of target object and electronic equipment | |
CN112734747B (en) | Target detection method and device, electronic equipment and storage medium | |
CN111062415B (en) | Target object image extraction method and system based on contrast difference and storage medium | |
CN117475253A (en) | Model training method and device, electronic equipment and storage medium | |
CN110619304A (en) | Vehicle type recognition method, system, device and computer readable medium | |
CN113688810B (en) | Target capturing method and system of edge device and related device | |
KR20160073490A (en) | System for assessment of safety level at construction site based on computer vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20190116 Address after: 300 000 No. 9 Xinhuan South Street, Haibin New Area, Tianjin Applicant after: Tencent Digital (Tianjin) Co.,Ltd. Address before: 518057 Tencent Building, No. 1 High-tech Zone, Nanshan District, Shenzhen City, Guangdong Province, 35 floors Applicant before: TENCENT TECHNOLOGY (SHENZHEN) Co.,Ltd. Applicant before: TENCENT CLOUD COMPUTING (BEIJING) Co.,Ltd. |
|
TA01 | Transfer of patent application right | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |