CN114202742A - Method and device for judging whether battery car enters elevator - Google Patents
Method and device for judging whether battery car enters elevator Download PDFInfo
- Publication number
- CN114202742A CN114202742A CN202210084228.8A CN202210084228A CN114202742A CN 114202742 A CN114202742 A CN 114202742A CN 202210084228 A CN202210084228 A CN 202210084228A CN 114202742 A CN114202742 A CN 114202742A
- Authority
- CN
- China
- Prior art keywords
- information
- battery car
- judging whether
- elevator
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a device for judging whether an accumulator car enters an elevator, and belongs to the field of image recognition. According to the invention, whether the foreground image enters the elevator or not is judged by calculating the motion tracks of the battery car and the human shape, so that the process that the battery car enters the elevator can be automatically identified, and the manpower is saved and the identification effect is high; the recognition algorithm is preposed in the monitoring equipment, so that the response can be quickly and efficiently made, and the occupied network resources are less; by performing mask processing on the background model, the foreground image is reserved, interference factors are reduced, and the identification accuracy of the system is improved. By selecting representative pixel points in the image information and judging the outline area of the foreground image, most interference factors are filtered, the system load of the micro-processing unit is reduced, and the requirement of the identification method on hardware equipment is reduced.
Description
Technical Field
The invention belongs to the field of image recognition, and particularly relates to a method and a device for judging whether an accumulator car enters an elevator.
Background
With the increase of the number of the battery cars, the problem of community management of the battery cars is increasingly prominent. Mainly embody in aspects such as the charging safety of storage battery car, a lot of storage battery car owner can directly ride the storage battery car home in for the electric motor car charges, charges in. Under the conditions of the existing battery and charging technology, the storage battery car has certain probability of fire when being charged. Once a fire happens to the battery car, the battery car can spread rapidly within a few seconds, so that safety accidents are easily caused, and the life and property safety of residents is endangered.
In the prior art, the data collected by the monitoring equipment is analyzed and transmitted to the control center, the automatic identification of the image is realized through the convolutional neural network algorithm of the control center, and then the result is fed back to the execution terminal. With the popularization of the algorithms, an embarrassing moment is met at present, and when the algorithms are applied, the network transmission system of the cell property needs to be upgraded and optimized, so that the commercial application of various electric vehicle identification methods is limited. In addition, the existing identification algorithm has a large occupied network resource, and when data transmission is unstable, the reaction delay of the execution unit can be caused.
Disclosure of Invention
The invention aims to provide a method, a device, a server and a readable storage medium for judging whether a battery car enters an elevator, so as to solve the problems related to the background technology.
Based on the technical problem, the invention provides a method and a device for judging whether a battery car enters an elevator, a server and a readable storage medium, and the method and the device comprise the following four aspects.
In a first aspect, the invention provides a method for judging whether a battery car enters an elevator, which is used for monitoring equipment, and the method comprises the following steps: establishing a first recognition model, wherein the first recognition model recognizes the battery car and the human shape; acquiring first information, wherein the first information is shot by the monitoring device; performing first analysis on the first information by the first recognition model to obtain a first analysis result; judging whether to perform a second analysis according to the first analysis result; the second analysis comprises calculating the motion trail of the battery car and the human shape; judging whether the battery car and the human figure enter the elevator or not according to the motion trail; and if so, executing a first instruction, wherein the first instruction comprises an alarm prompt.
Preferably, the method further comprises: establishing a preset first judgment model which is a rectangular judgment frame preset on the first information; extracting second information from the first information, wherein the second information comprises position information of the battery car and the human-shaped recognition frame; and judging the position change of the second information relative to the first judgment frame to obtain whether the battery car and the human figure enter the elevator.
Preferably, the method further comprises: establishing a first coordinate system, wherein the first coordinate system takes the lower left corner of the first information as an origin O and the horizontal direction as the ordinate of an abscissa tree; coordinates of a point at the upper left corner of the judgment frame are (X3, Y3), and coordinates of a point at the lower right corner of the judgment frame are (X4, Y4); the position information is coordinate values (X, Y) of the central points of the human shape and the electromobile recognition frame; acquiring changes of the figure and the Y coordinate value of the center point of the battery car identification frame in the second information; and judging whether the Y value is smaller than Y3 and Y4 in sequence, and if so, determining that the battery car and the human form enter the corridor.
Preferably, prior to the first analysis, the method further comprises: establishing a background model, wherein the background model is a background image which is always kept motionless in the first information; judging whether a foreground image appears in the background model; if yes, executing the next step; and judging whether the outline area of the foreground image is larger than a preset frame threshold value or not, and if so, executing a first analysis.
Preferably, the method further comprises: preprocessing the first information, wherein the preprocessing comprises selecting a predetermined number of pixel points in the image information according to a predetermined interval and acquiring pixel values of the pixel points; calculating the pixel difference value of the pixel points at the same position in the adjacent two frames of image information; judging whether the pixel difference value is smaller than a preset value, if so, taking the pixel point as a static point and taking a preset area around the pixel point as a static area; and accumulating the static point and the preset area to establish a background model.
Preferably, the method further comprises: preprocessing the first information, wherein the preprocessing comprises selecting a predetermined number of pixel points in the image information according to a predetermined interval and acquiring pixel values of the pixel points; calculating the pixel difference value of the pixel points at the same position in the adjacent two frames of image information; judging whether the pixel difference value is smaller than a preset value, if not, displaying a foreground image in the background model, and executing the next step; obtaining boundary pixels between the foreground image and the background model; and segmenting the first information along the boundary pixels to extract the foreground image.
Preferably, the method further comprises: processing the first information to obtain third information, wherein the third information is the first information for performing mask processing on a background model and reserving a foreground image; performing first analysis on the third information through the first recognition model to obtain the feature similarity of the third information; and judging whether the battery car and the human shape appear in the first information or not according to the feature similarity of the third information.
In a second aspect, the present invention further provides a device for determining whether a battery car enters an elevator, which is used for monitoring equipment, and the device includes:
the first preset unit is used for establishing a first recognition model, and the first recognition model recognizes the battery car and the human shape;
the first acquisition unit is used for acquiring first information, and the first information is shot by the monitoring device;
the first processing unit is used for carrying out first analysis on the first information through the first recognition model to obtain a first analysis result;
the first judging unit is used for judging whether to perform second analysis according to the first analysis result;
the second processing unit is used for carrying out second analysis, and the second analysis comprises calculation of the movement tracks of the battery car and the human shape;
the second judgment unit is used for judging whether the battery car and the human figure enter the elevator or not according to the motion track; and if so, executing a first instruction, wherein the first instruction comprises an alarm prompt.
In a third aspect, the server for identifying the entry of the battery car into the elevator comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to realize the steps of the method for identifying the entry of the battery car into the elevator.
In a fourth aspect, a computer-readable storage medium, wherein the program when executed by a processor implements the steps of the method for identifying entry of the battery car into an elevator.
Has the advantages that: the invention relates to a method and a device for judging whether an accumulator car enters an elevator, a server and a readable storage medium, wherein the method and the device can realize automatic identification on the process that the accumulator car enters the elevator by calculating the motion tracks of the accumulator car and a human shape and judging whether a foreground image enters the elevator, so that the manpower is saved and the identification effect is high; the recognition algorithm is preposed in the monitoring equipment, so that the response can be quickly and efficiently made, and the occupied network resources are less; by performing mask processing on the background model, the foreground image is reserved, interference factors are reduced, and the identification accuracy of the system is improved. By selecting representative pixel points in the image information and judging the outline area of the foreground image, most interference factors are filtered, the system load of the micro-processing unit is reduced, and the requirement of the identification method on hardware equipment is reduced.
Drawings
Fig. 1 is a schematic flow chart of an identification method for a battery car in embodiment 1 of the present invention.
Fig. 2 is a diagram of an identification device for a battery car in embodiment 2 of the present invention.
Fig. 3 is a schematic structural diagram of an exemplary electronic device in embodiment 3 of the present invention.
Description of reference numerals: the device comprises a first preset unit 11, a first acquisition unit 12, a first processing unit 13, a first judgment unit 14, a second processing unit 15, a second judgment unit 16, a bus 300, a receiver 301, a processor 302, a transmitter 303, a memory 304 and a bus interface 305.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the present invention. It will be apparent, however, to one skilled in the art, that the present invention may be practiced without one or more of these specific details. In other instances, well-known features have not been described in order to avoid obscuring the invention.
Example 1
As shown in the attached figure 1, the method for judging whether the battery car enters the elevator is used for monitoring equipment and comprises the following steps:
s100, establishing a first recognition model, wherein the first recognition model recognizes the electromobile and the human figure.
Specifically, the first identification model is an image identification technology, and the possibility that the battery car and the human-shaped combined image exist in any picture information can be calculated; in this embodiment, the identification model adopts a VoVNet network structure as a training backbone network, uses a centret algorithm as a core algorithm for identification and detection, and then trains a battery car and a human-shaped Pytorch model capable of detecting identification on a Pytorch deep learning frame at a PC end; the Pytrch model is converted into a Caffe model supported by Haisi 3516dv300 chip and is deployed on a Haisi board.
S200, acquiring first information, wherein the first information is shot by the monitoring device.
Specifically, the first information is video information in a specific area shot by the monitoring device, and the video information can be regarded as continuous multi-frame image information shot by the monitoring device according to a time sequence. The specific area is an area observed by the monitoring equipment and can be an elevator entrance or a necessary path to the elevator entrance. When the monitoring device is installed, the shooting angle needs to be reasonably adjusted so that the monitoring device covers the specific area as much as possible.
S300, performing first analysis on the first information through the first recognition model to obtain a first analysis result.
Specifically, the first analysis is to calculate the possibility that a battery car and a human-shaped combined image exist in the picture information through an image recognition technology. The first analysis result is that the first information is compared with the characteristic information prestored in the first recognition model to obtain the characteristic similarity; in this embodiment, a first recognition model for battery car recognition is preset, and then the first information is compared with the feature information prestored in the first recognition model to obtain the feature similarity of the first information.
However, in the existing image recognition technology, a candidate feature frame generally needs to be set in the first information, in order to improve the recognition accuracy, a convolution algorithm needs to be adopted to optimize the feature frame, remove an excessively large or excessively small frame to obtain an optimal feature frame, and then the feature information is recognized through a recognition model to reduce the problem of reduction of the recognition accuracy caused by external interference factors. However, since the recognition algorithm is used for monitoring equipment, the hardware requirement is high by adopting the algorithm to process through a micro-processing unit arranged on the monitoring equipment.
Thus, based on the above problem, the first information is preprocessed before the first analysis is performed. The preprocessing is an extraction process of the feature frame, and specifically, the preprocessing comprises the following steps: establishing a background model, wherein the background model is a background image which is always kept motionless in the first information; judging whether a foreground image appears in the background model; if yes, executing the next step; and judging whether the outline area of the foreground image is larger than a preset frame threshold value or not, and if so, executing a first analysis.
The background model is a background image which is kept still in the first information all the time, and the image information can be regarded as the background image by not changing the images on the continuous multi-frame image information in the video information. Microscopically, if the pixel value of the pixel point on the continuous multi-frame image information is kept unchanged or the pixel value is changed in a small range, the current image information is considered as the background model. The preset frame threshold is an area value and is related to the installation position of the monitoring equipment, the shooting angle and the resolution of the ray head; when a foreground image appears in the image information, whether the external outline area of the foreground image is larger than a frame threshold or not is judged, and if the external outline area of the foreground image is larger than the frame threshold, the foreground image is possibly a combined image of an electric vehicle and a human figure, and if the external outline area of the foreground image is not larger than the frame threshold, the foreground image is not possible to be the electric vehicle in the foreground image. So set up, can further avoid the interference of the foreground image of some obvious non-storage battery cars and personalities. For example, when a person or animal passes through a corridor, the identification algorithm is caused to perform ineffective identification, a large amount of computing resources are wasted, and the load of the micro-processing unit is increased. Therefore, by setting a frame threshold value and carrying out comparison and judgment, the outer contour area of the foreground image is compared with the area of the set frame, and the foreground image with the excessively small outer contour area is removed; therefore, noise is eliminated, and the problem of complex operation caused by the influence of external interference factors is solved.
In this embodiment, the method for establishing the background model includes the following steps: preprocessing the first information, wherein the preprocessing comprises selecting a predetermined number of pixel points in the image information according to a predetermined interval and acquiring pixel values of the pixel points; calculating the pixel difference value of the pixel points at the same position in the adjacent two frames of image information; judging whether the pixel difference value is smaller than a preset value, if so, taking the pixel point as a static point and taking a preset area around the pixel point as a static area; and accumulating the static point and the preset area to establish a background model.
Wherein, the preset distance and the preset number are set according to the probability of the foreground images appearing in the area, generally speaking, when the monitoring equipment is in a downward shooting mode, the probability of the foreground images appearing in the middle position is higher; when the monitoring equipment is used for side shooting, the probability that the foreground image appears at the lower position is higher, so that the area with higher probability that the foreground image appears in the image information is determined according to the installation angle of the monitoring equipment, the preset distance set in the area is smaller, and the preset number of pixel points is larger. The probability of the foreground image appearing in the image information may refer to the stored data of the monitoring device, specifically, the frequency of the foreground image appearing in the image information in a certain period of time before. The pixel value is a specific value given by a computer when the image information is digitized, represents the average brightness information of the pixel point, or the average reflection (transmission) density information of the small square, and the value of the pixel value is 0-255. The predetermined value is related to factors such as the resolution of the ray head, and in the embodiment, the predetermined value is 3-5.
In the existing background image recognition technology, it is generally required to calculate a pixel value of each pixel point in a continuous frame image, and then match each pixel value with previous image information to construct a complete background model. However, since the identification method is generally applied to areas such as residential buildings and the like, and the density of people flow is relatively low, most of the image information is in a static state, and if the existing background model construction technology is adopted, a large amount of saturated calculation exists, so that the load of the micro-processing unit is increased. In addition, when some obvious foreground images of non-battery cars and characters appear in the background model, such as the passing of birds, mice and other small-sized animals, even before tiny objects pass through the monitoring equipment, the identification method still executes the next step to intelligently identify the images, which also causes a great amount of computing resources to be wasted, aggravates the load of a micro-processing unit and is not favorable for long-term stable operation of the system. By adopting the method for establishing the background model, a predetermined number of representative pixel points are selected for calculation, so that the load of the microprocessing unit is greatly reduced; part of the interference factors may also be filtered.
Similarly, the foreground image identification method comprises the following steps: preprocessing the first information, wherein the preprocessing comprises selecting a predetermined number of pixel points in the image information according to a predetermined interval and acquiring pixel values of the pixel points; calculating the pixel difference value of the pixel points at the same position in the adjacent two frames of image information; judging whether the pixel difference value is smaller than a preset value, if not, displaying a foreground image in the background model, and executing the next step; and obtaining boundary pixels between the foreground image and the background model.
Specifically, when the image in the video information changes over the continuous multi-frame image information in the video information, whether the image is the background is judged. And if the pixel value of the pixel point on the continuous multi-frame image information is kept unchanged or the pixel value is changed within a small range, the foreground image is considered to appear in the current image information. The foreground image specific acquisition method comprises the following steps: preprocessing the continuous image information to obtain gray information and gradient information of the image; calculating the pixel value of each pixel point in the continuous frame image, then calculating the pixel difference value of the pixel point at the same position in the adjacent two frames of image information, and when the pixel difference value is greater than a preset value, considering the pixel point as a moving point; and accumulating the motion points to construct a complete foreground image.
The above-described acquisition method can acquire the size and position of the foreground image in preparation. However, the same requirements for computer equipment are high, and the hardware system of the existing residential building property needs to be upgraded and modified. Therefore, in order to simplify the calculation method, the inventors further devised a method of confirming the foreground image by contour extraction. The foreground image is segmented by identifying boundary pixels between the foreground image and the background model. Specifically, on the basis of step 230, a further determination is made, when the pixel value of the pixel point is obtained, then the pixel difference value of the pixel point at the same position in the two adjacent frames of image information is calculated, and when the pixel difference value is greater than a predetermined value, the pixel point is considered as a moving point, that is, it is said that a foreground image appears in the background model; the method comprises the steps of generally forming an obvious boundary between a foreground image and a background model, wherein the boundary is formed by different pixel properties of two sides of the boundary, the pixel properties specifically comprise the resolution, the color temperature and the tone characteristics of pixels, analyzing boundary pixels on the boundary, then segmenting image units in the boundary pixels, and extracting the foreground image.
Based on the pretreatment, the invention further optimizes the second analysis, and the second analysis comprises the following steps: processing the first information to obtain third information, wherein the third information is the first information for performing mask processing on a background model and reserving a foreground image; performing first analysis on the third information through the first recognition model to obtain the feature similarity of the third information; and judging whether the battery car and the human shape appear in the first information or not according to the feature similarity of the third information. In the embodiment, a first identification model for battery car identification is preset, then mask processing is carried out on a background model in first image information, and a foreground image is reserved; and comparing the foreground image with the characteristic information prestored in the identification model, obtaining the characteristic similarity of the foreground part in each frame of image information, and if the similarity is more than 75%, determining that the battery car and the human figure appear in the foreground image, and executing the step 400. Otherwise, the foreground image is determined to have no battery car or human shape, and the step S100 is executed. By adopting the method, the mask processing can be directly carried out on the background model, so that the calculation amount of the microprocessing unit can be reduced, and the problem of reduced identification accuracy caused by external interference factors can be solved.
And S400, judging whether to perform second analysis according to the first analysis result.
Specifically, when the first analysis result is smaller than a preset value, the person is considered that the battery car appears in the first information, and otherwise, the battery car does not appear. Specifically, in this embodiment, when the feature similarity is greater than 75%, it is determined that the battery car and the human figure appear in the foreground image, and step 500 is executed. Otherwise, the foreground image is determined to have no battery car or human shape, and the step S100 is executed.
S500, the second analysis comprises calculating the motion tracks of the battery car and the human shape;
specifically, the second analysis includes the steps of: establishing a first coordinate system, wherein the first coordinate system takes the lower left corner of the first information as an origin O and the horizontal direction as the ordinate of an abscissa tree; establishing a preset first judgment model, wherein the first judgment model is a rectangular judgment box preset on the first information, the coordinates of the upper left corner point of the judgment box are (X3, Y3), and the coordinates of the lower right corner point of the judgment box are (X4, Y4); extracting second information from the first information, wherein the second information comprises position information of the electromobile and the human-shaped recognition frame, and the position information is coordinate values (X, Y) of the central points of the human-shaped recognition frame and the electromobile recognition frame; acquiring changes of the figure and the Y coordinate value of the center point of the battery car identification frame in the second information; and judging whether the Y value is smaller than Y3 and Y4 in sequence, and if so, determining that the battery car and the human form enter the corridor.
In this embodiment, the second analysis includes the following steps:
s510, analyzing and identifying the ID and the coordinate values of the detected object, setting the human shape ID to be 0, the coordinate values of the human shape frame to be (X1, Y1, W1 and H1), the ID of the battery car to be 1 and the coordinate values to be (X2, Y2, W2 and H2). And the lower left corner of the picture is taken as an origin O, and X and Y are coordinates of the upper left corner of the identification frame, namely an abscissa and an ordinate relative to the origin O. W and H are the length and width of the identification box, respectively.
S520, calculating coordinate values of the central points of the human figure recognition frame and the battery car recognition frame respectively through the X, Y, W and H values;
s530, setting a preset rectangular frame at a corridor opening in a monitoring equipment picture, and respectively setting coordinates of upper left corner points and lower right corner points of the preset rectangular frame as (X3, Y3) and (X4, Y4);
s540, when the electromobile and the human figure are identified, the change of the Y coordinate values of the central points of the human figure and the electromobile identification frame in the continuous frames is analyzed and compared, the movement tracks of the human figure and the electromobile identification frame can be obtained by combining the X values, and then whether the electromobile and the human enter a passageway or exit the passageway is judged;
and S550, comparing and analyzing the coordinate value Y of the central point of the human-shaped and battery car identification frame with the preset rectangular frame Y3 and Y4 values, and determining that the electric car enters the corridor when the Y value is less than Y3 and Y4 in sequence.
In this embodiment, the method for determining the identification frame includes the following steps: when the battery car and the person are determined to exist in a certain frame of image information in the step 300, calculating the closed contour of the battery car and person combined graph in the subsequent continuous multi-frame image information, wherein the information in the closed contour can be understood as a part of the battery car and the person.
In this embodiment, the monitoring device is a side shooter, and therefore the contour identification method includes the following steps:
s511, taking the foreground image obtained in the step S300 as a first frame image; obtaining face features and joint features of a human body of the foreground image, front and rear wheel features of the battery car, seat plate features and car handle features, and obtaining a closed contour of the foreground image through a delineation method; further, coordinate values of the human-shaped frame of the first frame image are (X1, Y1, W1 and H1), and coordinate values of the battery car frame are (X2, Y2, W2 and H2);
s512, recording a plurality of continuous frame images behind the first frame image as an Nth frame image; obtaining the facial features and the joint features of the Nth frame image by comparing the facial features and the joint features of the first frame image;
s513, connecting all the outermost joint bending parts to form the body contour of the human body; identifying ear features of a human body, taking the ear features as the center of a face, determining the position of the face, and obtaining the length of the face by determining the distance from the chin features to the ear features as the distance from the chin to the ears is usually half of the length of the face; obtaining the face width by determining the distance from the anterior nose characteristic value to the ear characteristic, determining an ellipse according to the face length and width, and connecting the ellipse with the body contour to obtain the body contour;
s514, because the contour of the middle position of the battery car is overlapped with the contour of a human body, the rear contour of the battery car is obtained by connecting the front wheel characteristic, the handle characteristic and the contour of the human body; similarly, the characteristics of the front wheels, the characteristics of the seat plate and the human body profile are connected to obtain the rear profile of the battery car; and finally, connecting the characteristics of the front wheel and the rear wheel so as to determine the outline of the battery car and human body combined image.
S515, the coordinate values of the human-shaped frame of the N-th frame image are (X1, Y1, W1, H1), and the coordinate values of the battery car frame are (X2, Y2, W2, H2)
In the foreground image obtained from the foreground image obtained in the step S300, the key features of the battery car and the human figure image can be obtained, and if similar extraction steps are still adopted in the subsequent frame image, the calculation amount is increased undoubtedly, so that the contour extraction is performed on the foreground image of the subsequent frame image on the basis of the step S300, and the system load for identifying the battery car and the human body combined image is reduced.
S600, judging whether the battery car and the human figure enter an elevator or not according to the motion track; and if so, executing a first instruction, wherein the first instruction comprises an alarm prompt.
Specifically, the first instruction includes but is not limited to an alarm prompt, and for those skilled in the art, the first instruction may also be in other forms, such as a snapshot, a stop of the elevator, a non-closing of the elevator door, a light alarm, and a sound alarm.
Example 2
Based on the same inventive concept as the method for judging whether the battery car enters the elevator in the embodiment 1, the invention also provides a method and a device for judging whether the battery car enters the elevator, as shown in fig. 2, the device comprises:
the first preset unit 11 is used for establishing a first recognition model, and the first recognition model recognizes the battery car and the human shape;
a first obtaining unit 12, configured to obtain first information, where the first information is shot by the monitoring device;
the first processing unit 13 is used for performing first analysis on the first information by using the first recognition model to obtain a first analysis result;
a first judging unit 14, configured to judge whether to perform a second analysis according to the first analysis result;
the second processing unit 15 is used for performing second analysis, wherein the second analysis comprises calculation of the movement tracks of the battery car and the human shape;
the second judging unit 16 is used for judging whether the battery car and the human figure enter the elevator or not according to the motion track; and if so, executing a first instruction, wherein the first instruction comprises an alarm prompt.
The device, still include:
the second preset unit is used for establishing a first coordinate system, wherein the first coordinate system takes the lower left corner of the first information as an origin O, and the horizontal direction is the ordinate of the abscissa tree;
a third preset unit, configured to establish a predetermined first judgment model, where the first judgment model is a rectangular judgment box preset on the first information, and coordinates of an upper left corner point of the judgment box are (X3, Y3), and coordinates of a lower right corner point of the judgment box are (X4, Y4);
the third processing unit is used for extracting second information from the first information, the second information comprises position information of the electromobile and the human-shaped recognition frame, and the position information is coordinate values (X, Y) of the central points of the human-shaped recognition frame and the electromobile recognition frame;
and the third judgment unit is used for judging whether the position change of the second information relative to the first judgment frame is achieved by judging whether the Y value is smaller than Y3 and Y4 in sequence, and whether the battery car and the human figure enter the elevator is obtained.
The device, still include:
the fourth processing unit is used for establishing a background model, wherein the background model is a background image which is kept still in the first information all the time;
a fourth judging unit, configured to judge whether a foreground image appears in the background model; if yes, executing the next step;
and the fifth judging unit is used for judging whether the outline area of the foreground image is larger than a preset frame threshold value or not, and if so, executing the first analysis.
The device, still include:
the fifth processing unit is used for preprocessing the first information, wherein the preprocessing comprises the steps of selecting a preset number of pixel points in the image information according to a preset interval and acquiring pixel values of the pixel points;
the sixth processing unit is used for calculating the pixel difference value of the pixel point at the same position in the adjacent two frames of image information;
a fifth judging unit, configured to judge whether the pixel difference is smaller than a predetermined value, if so, the pixel is a stationary point, and a predetermined area around the pixel is a stationary area;
and the seventh processing unit is used for accumulating the static point and the preset area to establish a background model.
The device, still include:
a fifth judging unit, configured to judge whether the pixel difference is smaller than a predetermined value, and if not, perform the next step when a foreground image appears in the background model;
the eighth processing unit is used for obtaining boundary pixels between the foreground image and the background model;
and the ninth processing unit is used for segmenting the first information along the boundary pixels and extracting the foreground image.
The device, still include:
a tenth processing unit, configured to process the first information to obtain third information, where the third information is first information obtained by performing mask processing on a background model and reserving a foreground image;
the eleventh processing unit is used for carrying out first analysis on the third information through the first recognition model to obtain the feature similarity of the third information;
and the sixth judging unit judges whether the battery car and the human shape appear in the first information or not according to the characteristic similarity of the third information.
Various changes and specific examples of the method for judging whether the battery car enters the elevator in the first embodiment are also applicable to the device for judging whether the battery car enters the elevator in the first embodiment.
Example 3
Based on the same inventive concept as the method for determining whether the battery car enters the elevator in the foregoing embodiment, the present invention further provides a server for determining whether the battery car enters the elevator, as shown in fig. 3, fig. 3 is an exemplary electronic device in embodiment 3, and includes a memory 304, a processor 302, and a computer program stored on the memory 304 and capable of running on the processor 302, and when the processor 302 executes the program, the steps of any one of the methods for remote vital sign monitoring described above are implemented.
Where in fig. 3 a bus architecture (represented by bus 300), bus 300 may include any number of interconnected buses and bridges, bus 300 linking together various circuits including one or more processors, represented by processor 302, and memory, represented by memory 304. The bus 300 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 305 provides an interface between the bus 300 and the receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, i.e., a transceiver, providing a means for communicating with various other apparatus over a transmission medium.
The processor 302 is responsible for managing the bus 300 and general processing, and the memory 304 may be used for storing data used by the processor 302 in performing operations.
Example 4
Based on the same inventive concept as the method for determining whether the battery car enters the elevator in the foregoing embodiment, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which when executed by a processor, implements the steps of: establishing a first recognition model, wherein the first recognition model recognizes the battery car and the human shape; acquiring first information, wherein the first information is shot by the monitoring device; performing first analysis on the first information by the first recognition model to obtain a first analysis result; judging whether to perform a second analysis according to the first analysis result; the second analysis comprises calculating the motion trail of the battery car and the human shape; judging whether the battery car and the human figure enter the elevator or not according to the motion trail; and if so, executing a first instruction, wherein the first instruction comprises an alarm prompt.
One or more technical solutions in the embodiments of the present invention at least have one or more of the following technical effects: the movement tracks of the battery car and the human shape are calculated, whether the foreground image enters the elevator or not is judged, automatic identification can be achieved in the process that the battery car enters the elevator, labor is saved, and the identification effect is high; the recognition algorithm is preposed in the monitoring equipment, so that the response can be quickly and efficiently made, and the occupied network resources are less; by performing mask processing on the background model, the foreground image is reserved, interference factors are reduced, and the identification accuracy of the system is improved. By selecting representative pixel points in the image information and judging the outline area of the foreground image, most interference factors are filtered, the system load of the micro-processing unit is reduced, and the requirement of the identification method on hardware equipment is reduced.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. The invention is not described in detail in order to avoid unnecessary repetition.
Claims (10)
1. A method for judging whether an accumulator car enters an elevator is used for monitoring equipment, and comprises the following steps:
establishing a first recognition model, wherein the first recognition model is used for recognizing the battery car and the human shape;
acquiring first information, wherein the first information is shot by the monitoring device;
performing first analysis on the first information by the first recognition model to obtain a first analysis result;
judging whether to perform a second analysis according to the first analysis result;
the second analysis comprises calculating the motion trail of the battery car and the human shape;
judging whether the battery car and the human figure enter the elevator or not according to the motion trail; and if so, executing a first instruction, wherein the first instruction comprises an alarm prompt.
2. The method for judging whether the battery car enters the elevator or not according to claim 1, further comprising the following steps:
establishing a preset first judgment model which is a rectangular judgment frame preset on the first information;
extracting second information from the first information, wherein the second information comprises position information of the battery car and the human-shaped recognition frame;
and judging the position change of the second information relative to the first judgment frame to obtain whether the battery car and the human figure enter the elevator.
3. The method for judging whether the battery car enters the elevator or not according to claim 2, further comprising the following steps:
establishing a first coordinate system, wherein the first coordinate system takes the lower left corner of the first information as an origin O and the horizontal direction as the ordinate of an abscissa tree;
coordinates of a point at the upper left corner of the judgment frame are (X3, Y3), and coordinates of a point at the lower right corner of the judgment frame are (X4, Y4);
the position information is coordinate values (X, Y) of the central points of the human shape and the electromobile recognition frame;
acquiring changes of the figure and the Y coordinate value of the center point of the battery car identification frame in the second information;
and judging whether the Y value is smaller than Y3 and Y4 in sequence, and if so, determining that the battery car and the human form enter the corridor.
4. The method for determining whether the battery car enters the elevator according to claim 1, wherein before the first analysis, the method further comprises:
establishing a background model, wherein the background model is a background image which is always kept motionless in the first information;
judging whether a foreground image appears in the background model; if yes, executing the next step;
and judging whether the outline area of the foreground image is larger than a preset frame threshold value or not, and if so, executing a first analysis.
5. The method for judging whether the battery car enters the elevator or not according to claim 4, wherein the method further comprises the following steps:
preprocessing the first information, wherein the preprocessing comprises selecting a predetermined number of pixel points in the image information according to a predetermined interval and acquiring pixel values of the pixel points;
calculating the pixel difference value of the pixel points at the same position in the adjacent two frames of image information;
judging whether the pixel difference value is smaller than a preset value, if so, taking the pixel point as a static point and taking a preset area around the pixel point as a static area;
and accumulating the static point and the preset area to establish a background model.
6. The method for judging whether the battery car enters the elevator or not according to claim 4, wherein the method further comprises the following steps:
preprocessing the first information, wherein the preprocessing comprises selecting a predetermined number of pixel points in the image information according to a predetermined interval and acquiring pixel values of the pixel points;
calculating the pixel difference value of the pixel points at the same position in the adjacent two frames of image information;
judging whether the pixel difference value is smaller than a preset value, if not, displaying a foreground image in the background model, and executing the next step;
obtaining boundary pixels between the foreground image and the background model;
and segmenting the first information along the boundary pixels to extract the foreground image.
7. The method for judging whether the battery car enters the elevator or not according to claim 4, wherein the method further comprises the following steps:
processing the first information to obtain third information, wherein the third information is the first information for performing mask processing on a background model and reserving a foreground image;
performing first analysis on the third information through the first recognition model to obtain the feature similarity of the third information;
and judging whether the battery car and the human shape appear in the first information or not according to the feature similarity of the third information.
8. The utility model provides a whether storage battery car gets into judgement device of elevator which characterized in that for supervisory equipment, the device includes:
the first preset unit is used for establishing a first recognition model, and the first recognition model recognizes the battery car and the human shape;
the first acquisition unit is used for acquiring first information, and the first information is shot by the monitoring device;
the first processing unit is used for carrying out first analysis on the first information through the first recognition model to obtain a first analysis result;
the first judging unit is used for judging whether to perform second analysis according to the first analysis result;
the second processing unit is used for carrying out second analysis, and the second analysis comprises calculation of the movement tracks of the battery car and the human shape;
the second judgment unit is used for judging whether the battery car and the human figure enter the elevator or not according to the motion track; and if so, executing a first instruction, wherein the first instruction comprises an alarm prompt.
9. A server for identifying entry of an electric vehicle into an elevator, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 7 when executing the program.
10. A computer-readable storage medium, characterized in that the program is adapted to carry out the steps of the method of any one of claims 1-7 when executed by a processor.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110148702 | 2021-02-03 | ||
| CN2021101487024 | 2021-02-03 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114202742A true CN114202742A (en) | 2022-03-18 |
Family
ID=80658935
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210084228.8A Withdrawn CN114202742A (en) | 2021-02-03 | 2022-01-23 | Method and device for judging whether battery car enters elevator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114202742A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114803758A (en) * | 2022-04-26 | 2022-07-29 | 浙江科技学院 | Battery car identification and elevator control method and system for community elevator |
| CN117007101A (en) * | 2023-09-28 | 2023-11-07 | 广东星云开物科技股份有限公司 | Vehicle monitoring method, device, electronic equipment and storage medium |
| CN117372968A (en) * | 2023-12-07 | 2024-01-09 | 深圳市智慧城市通信有限公司 | Electric vehicle home-entering monitoring method based on Internet of things |
-
2022
- 2022-01-23 CN CN202210084228.8A patent/CN114202742A/en not_active Withdrawn
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114803758A (en) * | 2022-04-26 | 2022-07-29 | 浙江科技学院 | Battery car identification and elevator control method and system for community elevator |
| CN114803758B (en) * | 2022-04-26 | 2023-12-15 | 浙江科技学院 | Battery car identification and elevator control method and system for district elevator |
| CN117007101A (en) * | 2023-09-28 | 2023-11-07 | 广东星云开物科技股份有限公司 | Vehicle monitoring method, device, electronic equipment and storage medium |
| CN117007101B (en) * | 2023-09-28 | 2023-12-26 | 广东星云开物科技股份有限公司 | Vehicle monitoring method, device, electronic equipment and storage medium |
| CN117372968A (en) * | 2023-12-07 | 2024-01-09 | 深圳市智慧城市通信有限公司 | Electric vehicle home-entering monitoring method based on Internet of things |
| CN117372968B (en) * | 2023-12-07 | 2024-03-22 | 深圳市智慧城市通信有限公司 | Electric vehicle home-entering monitoring method based on Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114202742A (en) | Method and device for judging whether battery car enters elevator | |
| CN109522793B (en) | Method for detecting and identifying abnormal behaviors of multiple persons based on machine vision | |
| US11354901B2 (en) | Activity recognition method and system | |
| US10007850B2 (en) | System and method for event monitoring and detection | |
| US20170286780A1 (en) | Method and system for calculating passenger crowdedness degree | |
| CN111860274B (en) | Traffic police command gesture recognition method based on head orientation and upper half skeleton characteristics | |
| CN106372576A (en) | Deep learning-based intelligent indoor intrusion detection method and system | |
| CN106241534B (en) | More people's boarding abnormal movement intelligent control methods | |
| US20220366697A1 (en) | Image processing method and apparatus, electronic device and storage medium | |
| CN111553214B (en) | Method and system for detecting smoking behavior of driver | |
| CN105868690A (en) | Method and apparatus for identifying mobile phone use behavior of driver | |
| CN111160175A (en) | Intelligent pedestrian violation behavior management method and related product | |
| CN114202711A (en) | Intelligent monitoring method, device and system for abnormal behaviors in train compartment | |
| CN112016435A (en) | Abnormal behavior detection method and device, electronic equipment and storage medium | |
| CN112614260A (en) | Intelligent security system based on face recognition and positioning | |
| CN116443682B (en) | Intelligent elevator control system | |
| CN114359839A (en) | Method and system for identifying entrance of electric vehicle into elevator | |
| CN112580633B (en) | Public transport passenger flow statistics device and method based on deep learning | |
| CN115713721A (en) | Behavior posture recognition method based on dual-channel video collaborative perception | |
| WO2022267266A1 (en) | Vehicle control method based on visual recognition, and device | |
| CN100385947C (en) | System of identifying image of human bodies | |
| CN116030404A (en) | Artificial intelligence-based construction and safety monitoring method for electronic warning fence of operation area | |
| CN113177453A (en) | Intelligent monitoring system for intelligent community based on image recognition | |
| CN104751189A (en) | Pre-assembled dual-video analysis based real-time counting device and method of boarding and alighting bus passengers | |
| CN115215177A (en) | Intelligent identification system and method for elevator lifting |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220318 |
|
| WW01 | Invention patent application withdrawn after publication |