CN114743140A - Fire fighting access occupation identification method and device based on artificial intelligence technology - Google Patents

Abstract

The invention provides a fire fighting access occupation identification method and device based on an artificial intelligence technology, wherein the identification method comprises the following steps: acquiring video information in a monitoring area corresponding to a fire fighting channel, and performing video frame segmentation according to the video information to acquire a plurality of frames of image data; based on a preset target detection system, judging whether the fire fighting channel is occupied or not according to the image data, and if so, determining an occupied object; and determining corresponding occupied time according to the occupied object, judging whether the occupied time is greater than a preset alarm threshold value, and if so, executing alarm processing. The invention can improve the efficiency and accuracy of the fire fighting access occupation detection by detecting the fire fighting access occupation object and occupation time.

Description

Fire fighting access occupation identification method and device based on artificial intelligence technology

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to a fire fighting access occupation identification method and device based on artificial intelligence technology.

Background

At present, fire fighting access has important evacuation or remediation function when an emergency occurs, although the regions are prohibited from being occupied by private persons by law, because the probability of the emergency occurrence is low, the regions are in an idle state at ordinary times, the importance of the regions is often ignored, the existing common fire fighting access occupancy detection method still depends on citizen reporting, video monitoring patrol and the like, the method depends on manual work, and the response effect is poor, the invention patent CN202111455510 discloses a fire fighting access occupancy identification method and device based on an artificial intelligence technology, which directly acquires the fire fighting access occupancy object by acquiring video data in real time, inputs the video data to a trained classification identification model, and finally acquires the detection result of the fire fighting access occupancy, by the method, the requirement on hardware is strict, the calculation complexity is high, the fire fighting access occupancy object is directly acquired, but the occupancy result is not processed, the whole process lacks integrity and subsequent execution force, and the efficiency of fire fighting access management is reduced.

Disclosure of Invention

The invention provides a fire fighting access occupation identification method and device based on an artificial intelligence technology, which are used for solving the problems that an occupied object cannot be clearly determined and alarm processing cannot be carried out according to occupied time.

A fire fighting access occupation identification method based on an artificial intelligence technology comprises the following steps:

acquiring video information in a monitoring area corresponding to a fire fighting channel, and performing video frame segmentation according to the video information to acquire a plurality of frames of image data;

judging whether the fire fighting channel is occupied or not according to the image data based on a preset target detection system, and if so, determining an occupied object;

and determining corresponding occupation time according to the occupation object, judging whether the occupation time is greater than a preset alarm threshold value, and if so, executing alarm processing.

As an embodiment of the present invention: the method for acquiring the video information in the monitoring area corresponding to the fire fighting channel, segmenting the video frame according to the video information, and acquiring a plurality of frames of image data comprises the following steps:

acquiring position information of a fire fighting access, judging whether a monitoring device exists in the fire fighting access area according to the position information, and determining a judgment result;

when the judgment result shows that the monitoring device exists in the fire fighting access area, acquiring video information of the monitoring area;

according to the video information of the monitoring area, carrying out area division on the video information, and determining an area division result; wherein the region division result comprises: fire-fighting vehicle driving areas, other areas;

and carrying out segmentation processing on the video information corresponding to the fire fighting vehicle driving area to obtain a plurality of frames of image data of the fire fighting vehicle driving area.

As an embodiment of the present invention: the method comprises the following steps of judging whether the fire fighting channel is occupied or not according to the image data based on a preset target detection system, and if so, determining an occupied object, including:

performing image recognition on the image data of the monitoring area corresponding to the fire fighting channel to obtain an image recognition result; wherein the image recognition result comprises: the fire fighting channel is occupied and the fire fighting channel is not occupied;

when the image recognition result shows that the driving area of the fire fighting vehicle is occupied, determining an occupied object occupying the driving area based on a preset target detection system; wherein the occupancy object comprises: automotive, non-automotive, object.

As an embodiment of the invention: determining corresponding occupation time according to the occupation object, judging whether the occupation time is greater than a preset alarm threshold value, and if so, executing alarm processing, wherein the alarm processing comprises the following steps:

setting a fire fighting channel time occupation threshold according to different occupied objects, and sending occupation information corresponding to a fire fighting channel to a preset video monitoring platform when the occupied time of the occupied object reaches a preset time occupation threshold according to the occupied time of the occupied object; wherein the occupancy information comprises: occupied objects, occupied time.

As an embodiment of the present invention: the acquiring determines corresponding occupation time according to the occupation object, judges whether the occupation time is greater than a preset alarm threshold value, and if so, executes alarm processing, and further includes:

when the occupied time of the occupied object is larger than a preset time occupation threshold, acquiring a target detection result of the occupied object;

when the occupied object is determined to be the motor vehicle, identifying and acquiring license plate information of the motor vehicle, acquiring vehicle owner information and a telephone number of the motor vehicle based on a preset big data information platform, and intelligently sending a reminding short message;

when the occupied object is a non-motor vehicle, acquiring an initial occupied time point of the occupied object according to monitoring video information, acquiring facial image data of a corresponding owner of the non-motor vehicle according to the initial occupied time point, extracting facial features according to the facial image data, constructing a facial information matrix according to the facial features, acquiring a matching result of a target object according to the facial information matrix based on the big data information platform, determining personnel information and a telephone number of the target object when the matching result is greater than a preset matching threshold value, and intelligently sending a reminding short message;

when the occupied object is an object, acquiring an initial occupied time point of the occupied object according to monitoring video information, acquiring face information of the owner corresponding to the object according to the initial occupied time point, acquiring personnel information and a telephone number of the owner of the object based on the big data information platform, and intelligently sending a reminding short message.

As an embodiment of the invention: the method comprises the steps of obtaining video information in a monitoring area corresponding to a fire fighting channel, segmenting video frames according to the video information, obtaining a plurality of frames of image data, and further comprising: preprocessing the image data to obtain an image preprocessing result, wherein the preprocessing process comprises the following steps:

carrying out graying processing on the image data to obtain primary image data;

acquiring pixel point distribution corresponding to the primary image data, judging whether image noise exists in the primary image data according to the pixel point distribution result, and if so, performing noise reduction processing on the primary image data by adopting a spatial filtering method to acquire secondary image data; the spatial filtering method is used for calculating the gray value of a pixel point in image data through a template;

randomly sampling the secondary image data to obtain a signal discrete sample corresponding to the secondary image data, performing image reconstruction on the signal discrete sample by a nonlinear reconstruction method, and obtaining connected domain data information corresponding to the secondary image data according to an image reconstruction result; the nonlinear reconstruction method is used for obtaining low-frequency component loss of image data and performing high-frequency reconstruction aiming at the low-frequency component loss.

As an embodiment of the present invention: the method comprises the following steps that based on a preset target detection system, whether the fire fighting channel is occupied or not is judged according to the image data, if yes, an occupied object is determined, and the method further comprises the following steps:

acquiring fire station alarm information, determining the current location and target address of a fire truck according to the alarm information, planning a path according to the current location and target address of the fire truck, and determining a path planning result;

acquiring fire fighting channel setting information in the path according to the path planning result, acquiring monitoring information of a corresponding position according to the fire fighting channel setting information, judging whether a fire fighting channel is occupied according to the monitoring information, and determining a judgment result;

when the judgment result shows that the fire fighting channel is occupied in the path, acquiring the occupation information of the fire fighting channel; wherein the fire fighting access occupancy information comprises: the fire fighting access occupies the position and the object information;

and sending the fire fighting access occupation information to an emergency accident handling port.

As an embodiment of the present invention: the method for acquiring the video information in the monitoring area corresponding to the fire fighting channel, segmenting the video frame according to the video information, and acquiring a plurality of frames of image data comprises the following steps:

receiving fire fighting access occupation information uploaded by a mobile equipment port based on a preset big data information platform; the method for uploading information by the mobile equipment port comprises the following steps: wireless communication mode uploads, the phone reports and uploads, fire control passageway occupies information and includes: occupation address and occupation vehicle information;

performing data integration aiming at the fire fighting access occupation information to acquire sequence information occupied by the fire fighting access;

and acquiring vehicle license plate information occupying the fire fighting access according to the sequence information occupied by the fire fighting access, acquiring a contact way of a host according to the license plate information, and intelligently reminding according to the contact way.

As an embodiment of the present invention: the method comprises the steps of obtaining video information in a monitoring area corresponding to a fire fighting channel, segmenting video frames according to the video information, obtaining a plurality of frames of image data, and further comprising:

acquiring corresponding accident reasons and accident addresses according to alarm information of fire police, establishing a mapping relation between the accident addresses and the accident reasons, predicting the accident based on a preset big data information platform, and determining a predicted accident address;

acquiring corresponding monitoring video information and fire fighting channel distribution information in the accident address area according to the predicted accident address, acquiring the occupation situation of a fire fighting channel, and acquiring the contact way of people occupying the fire fighting channel according to the occupation situation of the fire fighting channel to carry out intelligent reminding;

and when the contact information of the person occupying the fire fighting access cannot be acquired, executing alarm processing, and sending the fire fighting access occupation information to an alarm port.

A fire fighting access occupation recognition device based on artificial intelligence technology comprises:

an image collector: the system comprises a monitoring area, a fire fighting channel and a video processing unit, wherein the monitoring area is used for acquiring video information in a monitoring area corresponding to the fire fighting channel, and performing video frame segmentation according to the video information to acquire a plurality of frames of image data;

fire-fighting access occupation detector: the system is used for judging whether the fire fighting channel is occupied or not according to the image data based on a preset target detection system, and if so, determining an occupied object;

fire control passageway occupies the alarm: and the device is used for determining corresponding occupied time according to the occupied object, judging whether the occupied time is greater than a preset alarm threshold value or not, and executing alarm processing if the occupied time is greater than the preset alarm threshold value.

The invention has the beneficial effects that: in this technical scheme at first carry out the video frame according to the video frame data in the monitoring video region and cut apart, acquire a plurality of frame image data, but not select whole video frame data, be favorable to reducing the complexity of calculation, improve the operating efficiency, secondly, through discerning fire control passageway occupation object, be favorable to occupying different processing methods of intelligence selection according to the difference, make the processing to fire control passageway occupation have more pertinence, report to the police to the occupation of fire control passageway through the mode of presetting the time threshold value at last, be favorable to reducing the speed and the efficiency that fire control passageway was handled, effectively avoid the condition that fire control passageway occupied to take place.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

fig. 1 is a schematic flow chart of a fire fighting access occupation identification method based on an artificial intelligence technology in an embodiment of the present invention;

FIG. 2 is a block diagram of a flow structure of a fire fighting access occupation identification method based on an artificial intelligence technology in an embodiment of the present invention;

fig. 3 is a schematic structural diagram of image preprocessing in a fire fighting access occupation identification method based on an artificial intelligence technology in the embodiment of the present invention;

fig. 4 is a schematic structural diagram of a fire fighting access occupation identification device based on an artificial intelligence technology in an embodiment of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.

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, and "plurality" means two or more unless specifically limited otherwise. 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.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Example 1:

the embodiment of the invention provides a fire fighting access occupation identification method based on an artificial intelligence technology, which comprises the following steps as shown in the attached drawings 1 and 2:

acquiring video information in a monitoring area corresponding to a fire fighting channel, and performing video frame segmentation according to the video information to acquire a plurality of frames of image data;

based on a preset target detection system, judging whether the fire fighting channel is occupied or not according to the image data, and if so, determining an occupied object;

determining corresponding occupation time according to the occupation object, judging whether the occupation time is greater than a preset alarm threshold value, and if so, executing alarm processing;

in one practical scenario: when the occupation of the fire fighting channel is detected, a traditional background difference and template-based vehicle detection method is adopted, the method is only suitable for detecting continuous vehicle video frames, the calculated amount of an algorithm is large in the mode, and the corresponding time complexity is high, so that the performance requirement on hardware equipment is strict, the relative maintenance cost is high, the characteristic-based vehicle detection method analyzes and detects the extracted characteristics of the vehicle, the detection rate is ideal under the condition of uniform illumination, the operation efficiency is high in the mode, but in the complex environment with obvious illumination change, the detection efficiency is obviously reduced, and the corresponding calculation complexity is high;

when the method is implemented and aiming at the occupation condition detection of the fire fighting access, the occupation condition of the fire fighting access area is detected, the occupied object and the occupation time are determined, the calculation efficiency is high, the calculation complexity is low, the adaptability to the change of the illumination condition is strong, the detailed outline of the occupied object is not required to be extracted, only the edge outline is required to be extracted, and the result is preliminarily determined;

in one specific embodiment, when determining whether the fire fighting access is occupied, training is first performed on the sample, assuming a given training sample (a)₁,b₁),…， (a_i,b_i) Where a represents the feature vector corresponding to the training sample，b_mClass label representing classification of object, m represents total number of samples, b_m∈[-1,1]Where-1 and 1 denote positive and negative samples, respectively, m is 1,2, …, i,

the weight values of the samples are initialized,

wherein σ_mRepresents the weight initialization result for m samples, p represents the number of positive samples, n represents the number of negative samples,

normalizing the weight subjected to initialization treatment:

and (3) outputting a classifier:

where T denotes the number of iterations, T1, 2, …, T, x denotes the sample value, h_t(x) The weak classification function aiming at the T-th iteration is represented, H (x) represents a strong classification function representation result added with the weak classification function, wherein the weak classification function in the formula represents the classification confidence coefficient of the weak classifier, the strong classification function consists of T weak classifiers, when the result of H (x) is more than 0, the classifier is represented to judge that the vehicle information exists in the sample data, and when the result of H (x) is less than 0, the classifier is represented to judge that the vehicle information does not exist in the sample data;

the beneficial effects of the above technical scheme are: in this technical scheme at first carry out the video frame according to the video frame data in the monitoring video region and cut apart, acquire a plurality of frame image data, but not select whole video frame data, be favorable to reducing the complexity of calculation, improve the operating efficiency, secondly, through discerning fire control passageway occupation object, be favorable to occupying different processing methods of intelligence selection according to the difference, make the processing to fire control passageway occupation have more pertinence, report to the police to the occupation of fire control passageway through the mode of presetting the time threshold value at last, be favorable to reducing the speed and the efficiency that fire control passageway was handled, effectively avoid the condition that fire control passageway occupied to take place.

Example 2:

in one embodiment, the acquiring video information in a monitoring area corresponding to a fire fighting access, performing video frame segmentation according to the video information, and acquiring a plurality of frames of image data includes:

acquiring position information of a fire fighting access, judging whether a monitoring device exists in the fire fighting access area according to the position information, and determining a judgment result;

when the judgment result shows that the monitoring device exists in the fire fighting access area, acquiring video information of the monitoring area;

according to the video information of the monitoring area, carrying out area division on the video information, and determining an area division result; wherein the region division result comprises: fire-fighting vehicle driving areas, other areas;

segmenting the video information corresponding to the fire fighting vehicle driving area to obtain a plurality of frames of image data of the fire fighting vehicle driving area;

in one practical scenario: when a monitoring device is detected to exist in the fire fighting channel, video information of a monitoring area is obtained, fire fighting channel occupation detection is directly carried out aiming at the video information, complete detection is carried out according to image data of each frame through the mode, and when real processing is carried out, the effective part is only a fire fighting vehicle driving area in the image data, if the complete image is subjected to occupation detection, the calculation complexity is improved, the difficulty of the effective area is easily increased by invalid areas of other areas, interference is easily caused, and the detection speed and efficiency are reduced;

when the method is implemented, after the monitoring video corresponding to the fire fighting channel is obtained, firstly, the video data is segmented, because the complexity and the accuracy for image processing are low and high, the memory occupied by the video processing is large, in addition, invalid areas in the image data, namely areas outside the fire fighting vehicle driving channel, are ignored, and the image processing is directly carried out on the fire fighting vehicle driving areas;

the beneficial effects of the above technical scheme are: according to the invention, video segmentation is carried out on video data to obtain a plurality of frames of image data, and fire fighting channel occupation detection is carried out on the image data, so that the complexity in the calculation process is reduced, and the calculation speed and efficiency are improved; in addition, the image data are subjected to region division, and only the effective driving region of the fire fighting channel is processed, so that the interference of the invalid region in the image data on the image processing is eliminated, and the accuracy of calculating the occupation condition of the fire fighting channel is improved.

Example 3:

in one embodiment, the determining, by the preset target detection system, whether the fire fighting channel is occupied according to the image data, and if so, determining an occupied object includes:

performing image recognition on the image data of the monitoring area corresponding to the fire fighting channel to obtain an image recognition result; wherein the image recognition result comprises: the fire fighting access is occupied and the fire fighting access is unoccupied;

when the image recognition result shows that the driving area of the fire fighting vehicle is occupied, determining an occupied object occupying the driving area based on a preset target detection system; wherein the occupancy object comprises: automotive, non-automotive, object;

in one practical scenario: when the fire fighting channel is detected, as long as the fire fighting channel is detected to contain objects, the fire fighting channel is judged to be occupied, wherein the fire fighting channel comprises walking people and bicycling people, the detection effect is not intelligent enough, and when the occupancy condition of the fire fighting channel is detected, if the fire fighting channel is detected, a telephone is used for reminding, and when other objects are detected to occupy the fire fighting channel, such as non-motor vehicles or objects, corresponding measures cannot be taken for stopping;

when the method is implemented, firstly, image analysis is carried out on image data of a monitoring area to obtain whether an object occupies an effective area in a fire fighting channel, when the object is detected to occupy the area in the fire fighting channel, the occupied object is further detected, and considering the situation that the occupied object has more and more complicated types, the method simply divides the types into: the method can preliminarily master the occupation condition of the fire fighting access, and provides an information basis for subsequent processing;

the beneficial effects of the above technical scheme are: according to the invention, whether the occupation condition occurs in the effective area in the fire fighting channel is firstly identified, when the occupation condition does not occur, the treatment is not carried out, only when the fire fighting channel is occupied, the subsequent calculation treatment is carried out, so that the calculation efficiency is favorably improved, the calculation complexity is reduced, secondly, the classification and the detection treatment are carried out aiming at the occupation object of the fire fighting channel, aiming at the occupation condition, the occupation behaviors in the monitoring area can be analyzed by utilizing the monitoring image and combining with an intelligent analysis algorithm, the image identification is respectively carried out on people, vehicles and objects, the behavior results of vehicle driving, pedestrian passing, object occupation, vehicle occupation, electric vehicle illegal parking and the like are distinguished, and the management efficiency of the fire fighting channel is favorably improved.

Example 4:

in one embodiment, the determining, according to the occupied object, a corresponding occupied time, determining whether the occupied time is greater than a preset alarm threshold, and if so, executing alarm processing, including:

setting a fire fighting channel time occupation threshold according to different occupied objects, and sending occupation information corresponding to a fire fighting channel to a preset video monitoring platform when the occupied time of the occupied object reaches a preset time occupation threshold according to the occupied time of the occupied object; wherein the occupancy information comprises: occupied objects, occupied time;

in one practical scenario: the occupation of the fire fighting access is detected by acquiring the video data, and when an object is detected in the video data, a result that the fire fighting access is occupied is made, wherein the result comprises pedestrians and running vehicles which are not required to be classified into the type of the occupied object, so that the detection efficiency of the occupation condition of the fire fighting access is low and most invalid detection results are easy to appear by the method;

when the method is implemented, corresponding occupation time thresholds are set for different objects, wherein the set initial occupation objects comprise: the method comprises the steps that different occupation time thresholds are set for different occupied objects of motor vehicles, non-motor vehicles and objects, a detection result whether a current fire fighting channel is occupied is made, information such as occupied time of the different occupied objects is obtained, and the occupation time of the fire fighting channel is further obtained;

the beneficial effects of the above technical scheme are: according to the invention, different time occupation thresholds are set for different occupied objects, so that whether the fire fighting channel is occupied or not can be judged according to the time of the occupied objects occupying the fire fighting channel, and the occupation accuracy of the fire fighting channel is further improved.

Example 5:

in one embodiment, the determining, according to the occupied object, a corresponding occupied time, determining whether the occupied time is greater than a preset alarm threshold, and if so, executing alarm processing, further includes:

when the occupied time of the occupied object is larger than a preset time occupation threshold, acquiring a target detection result of the occupied object;

when the occupied object is determined to be a motor vehicle, identifying and acquiring license plate information of the motor vehicle, acquiring vehicle owner information and a telephone number of the motor vehicle based on a preset big data information platform, and intelligently sending a reminding short message;

when the occupied object is a non-motor vehicle, acquiring an initial occupied time point of the occupied object according to monitoring video information, acquiring facial image data of a corresponding owner of the non-motor vehicle according to the initial occupied time point, extracting facial features according to the facial image data, constructing a facial information matrix according to the facial features, acquiring a matching result of a target object according to the facial information matrix based on the big data information platform, determining personnel information and a telephone number of the target object when the matching result is greater than a preset matching threshold value, and intelligently sending a reminding short message;

when the occupied object is an object, acquiring an initial occupied time point of the occupied object according to monitoring video information, acquiring face information of the owner corresponding to the object according to the initial occupied time point, acquiring personnel information and a telephone number of the owner of the object based on the big data information platform, and intelligently sending a reminding short message;

in one practical scenario: the occupation of the fire fighting channel is detected by acquiring the video data, and when an object is detected in the video data, a judgment result that the fire fighting channel is occupied is made, wherein the judgment result comprises pedestrians and running vehicles which are not required to be classified into the type of the occupied object, so that if the mode is adopted, the detection efficiency of the occupation condition of the fire fighting channel is low easily, and most invalid detection results are easy to appear;

when the method is implemented, the occupation time of the occupation object for the fire fighting channel is acquired, when the occupation time of the occupation object exceeds a set time occupation threshold value, the corresponding fire fighting channel is judged to be occupied, in addition, different time occupation threshold values are set for different occupation objects, and different reminding modes are set for different occupation opposite directions after the occupation object reaches the corresponding time threshold value;

in a specific embodiment, because the number of motor vehicles occupying the fire fighting passage is large, the accuracy of vehicle license plate recognition and detection in target detection occupied by the fire fighting passage is important, factors influencing the accuracy of license plate information detection such as vehicle parking angle, ambient light and the like are more, which can lead to fuzzification of the acquired license plate information and further influence the accurate acquisition of the license plate information,

step 1: constructing a similarity projection matrix aiming at the fuzzy license plate information, obtaining the feature vectors with the same dimensionality by mapping the fuzzy license plate information to the same feature representation space, supposing that the similarity projection matrix corresponding to the collected fuzzy license plate image data is Q,

the method comprises the following steps that A represents the length corresponding to collected fuzzy license plate image data, B represents the width corresponding to the collected fuzzy license plate image data, y (j, k) represents the projection distance corresponding to the collected fuzzy license plate image data in a three-dimensional space, j represents the abscissa corresponding to the collected fuzzy license plate image data in the three-dimensional space, and k represents the ordinate corresponding to the collected fuzzy license plate image data in the three-dimensional space;

step 2: according to the similarity projection matrix of the fuzzy license plate image data, obtaining local feature description of a license plate, obtaining local feature description information of low-frequency information and high-frequency information in the fuzzy license plate image data, dividing the high-frequency information into an information space, dividing the low-frequency information into a human cognitive space, using a region feature extraction method, carrying out feature description on the collected fuzzy license plate image data, and assuming that a local feature description result after feature extraction is:

wherein d represents a vector dimension value of the acquired fuzzy license plate image data, in general, d is 1, m represents the fuzzy license plate image data in the information space mode, n represents the fuzzy license plate image data in the human cognition space mode, and T represents_s() Representing the dimension element value of the local characteristic vector of the collected fuzzy license plate image data under the dimension d,

through the calculation of the steps, the local feature description information of the collected fuzzy license plate image data can be obtained, and a basis is further provided for calculating the feature similarity of the image;

and step 3: according to the local feature description result of the blurred license plate image data, a complete license plate information recognition result is obtained through a similarity measurement method, and a calculation formula of the similarity measurement method is assumed as follows:

wherein x is_m,nRepresenting the d-dimensional element value in the fuzzy license plate image data characteristics under different spatial modes; x is the number of_n,cRepresenting the value of the c-dimension element p in the fuzzy license plate image data under different spatial modes_cRepresenting the recognition result of the simulated license plate image data in the space mode, d representing the vector dimension value of the collected fuzzy license plate image data, wherein d is 1 in general, and x in general_m,nAnd x_n,cWhen all are less than 0, p_cIs 0 when x_m,nAnd x_n,cWhen both are greater than 0, p_cIs 1 when p_cAnd when the number of the images is 1, the similarity identification result aiming at the fuzzy license plate image data is valid.

The beneficial effects of the above technical scheme are: the method and the device are beneficial to improving the management efficiency of the occupation situation of the fire fighting channel by setting different time occupation thresholds aiming at different occupied objects, and are beneficial to improving the accuracy of license plate information acquisition and improving the reliability of the detection of the occupied objects of the fire fighting channel by detecting the fuzzy license plate information acquired from the monitoring video data.

Example 6:

in an embodiment, as shown in fig. 3, the acquiring video information in a monitoring area corresponding to a fire fighting access, performing video frame segmentation according to the video information, and acquiring a plurality of frames of image data further includes: preprocessing the image data to obtain an image preprocessing result, wherein the preprocessing process comprises the following steps:

carrying out graying processing on the image data to obtain primary image data;

acquiring pixel point distribution corresponding to the primary image data, judging whether image noise exists in the primary image data according to the pixel point distribution result, and if so, performing noise reduction processing on the primary image data by adopting a spatial filtering method to acquire secondary image data; the spatial filtering method is used for calculating the gray value of a pixel point in image data through a template;

randomly sampling the secondary image data to obtain a signal discrete sample corresponding to the secondary image data, performing image reconstruction on the signal discrete sample by a nonlinear reconstruction method, and obtaining connected domain data information corresponding to the secondary image data according to an image reconstruction result; the nonlinear reconstruction method is used for obtaining low-frequency component loss of image data and performing high-frequency reconstruction aiming at the low-frequency component loss;

in one practical scenario: considering the reason of economic cost, the resolution and image quality of the monitoring video data can not reach a high level, therefore, noise exists in the image frame data, and due to the change of light, the processing efficiency in the image processing link is low, the accuracy and efficiency occupied by the fire fighting access are reduced, and because the data collected by the camera device is mostly a color image, namely a 24-bit RGB true color image, the color information quantity contained in the color image is large, which not only occupies a large amount of memory space, but also consumes a large amount of system resources, greatly reduces the processing speed of the image, and the collected vehicle information can also have surrounding scenes with similar colors to the vehicle except the vehicle, the accuracy of system detection is also reduced, in addition, because the image collecting device, communication transmission and storage device are imperfect, the collected video or image is easily subjected to noise interference of different degrees, for example: in the process of transmitting an image through a medium, noise may be generated due to interference of external factors; in real life, most road vehicle images captured by image acquisition equipment are images containing noise, and the noise forms various spots in the images, so that the gray value of certain pixel points is changed, and the quality of the images is greatly reduced;

in the technical scheme, firstly, image graying is carried out on the image data, the memory occupied by the image can be reduced, denoising processing is carried out on the image subjected to the gray processing, before subsequent image processing and vehicle algorithm identification are carried out, a proper noise elimination algorithm needs to be selected according to the noise interference degree of an original image to improve a degraded image, and data information of an effective area in the image data is obtained by carrying out image reconstruction on secondary image data;

the beneficial effects of the above technical scheme are: according to the invention, after the color image is grayed, not only can the interference information in the image data be reduced, the memory space occupied by the image is reduced, and the calculation speed is favorably improved, but also the operation efficiency of the system is greatly improved, and the requirement of the real-time performance of the system can be well met.

Example 7:

in one embodiment, the system for detecting a fire fighting channel based on a preset target, which determines whether the fire fighting channel is occupied according to the image data, and if yes, determines an occupied object, further includes:

acquiring fire station alarm information, determining the current location and target address of a fire truck according to the alarm information, planning a path according to the current location and target address of the fire truck, and determining a path planning result;

acquiring fire fighting channel setting information in the path according to the path planning result, acquiring monitoring information of a corresponding position according to the fire fighting channel setting information, judging whether a fire fighting channel is occupied according to the monitoring information, and determining a judgment result;

acquiring fire fighting channel occupation information when the judgment result shows that the situation that the fire fighting channel is occupied exists in the path; wherein the fire fighting access occupancy information comprises: the fire fighting access occupies the position and the object information;

sending the fire fighting channel occupation information to an emergency accident handling port;

in one practical scenario: after receiving the alarm-out instruction, the fire station carries out path planning by the system, and the system cannot be connected with the system occupied by the fire channel in the process, so that the situation that the fire channel is occupied and then real-time processing is carried out in the process occurs, the processing speed is lagged, and serious consequences can be caused;

when the method is implemented, the positioning and target address of a fire fighting fleet is obtained according to fire fighting alarm information, path planning is carried out according to the positioning and target address of the fire fighting fleet, a path planning result is determined, corresponding monitoring information is obtained according to fire fighting channel setting information, whether the occupation condition of a fire fighting channel exists or not is obtained, when the judgment result shows that the situation that the fire fighting channel is occupied exists in the path, the occupation condition is immediately sent to an alarm port or an emergency accident handling port, and preprocessing work is carried out before the fire fighting vehicle arrives;

the beneficial effects of the above technical scheme are: through taking the detecting system with the fire control passageway and uniting with the system of giving an police, be favorable to improving the fire team and give an police smooth degree, avoid appearing the condition that the fire control passageway is taken after arriving the scene, improve the managerial efficiency and the rescue efficiency of system.

Example 8:

in one embodiment, the acquiring video information in a monitoring area corresponding to a fire fighting access, performing video frame segmentation according to the video information, and acquiring a plurality of frames of image data includes:

receiving fire fighting access occupation information uploaded by a mobile equipment port based on a preset big data information platform; the method for uploading information by the mobile equipment port comprises the following steps: wireless communication mode uploads, the phone reports and uploads, fire control passageway occupies information and includes: occupation address and occupation vehicle information;

performing data integration aiming at the fire fighting access occupation information to acquire sequence information occupied by the fire fighting access;

acquiring vehicle license plate information occupying the fire fighting access according to the sequence information occupied by the fire fighting access, acquiring a contact way of a host according to the license plate information, and carrying out intelligent reminding according to the contact way;

in one practical scenario: the condition that the fire fighting channel is occupied is obtained through online reporting and checking of management personnel by citizens, the labor cost is high, the condition that the fire fighting channel is occupied cannot be integrally mastered, the management mode is passive, and the efficiency is low;

when the method is implemented, the fire fighting access occupation information uploaded by a mobile equipment port is received through a preset big data information platform, data integration is carried out on the fire fighting access occupation information, sequence information and vehicle license plate information occupied by the fire fighting access are obtained, and intelligent reminding is carried out;

the beneficial effects of the above technical scheme are: the method and the system combine reporting results of citizens with an online fire fighting channel occupation platform, so that the management efficiency of the fire fighting channel occupation condition is improved.

Example 9:

in one embodiment, the acquiring video information in a monitoring area corresponding to a fire fighting access, performing video frame segmentation according to the video information, and acquiring a plurality of frames of image data further includes:

acquiring corresponding accident reasons and accident addresses according to alarm information of fire police, establishing a mapping relation between the accident addresses and the accident reasons, predicting the accident based on a preset big data information platform, and determining a predicted accident address;

acquiring corresponding monitoring video information and fire fighting channel distribution information in the accident address area according to the predicted accident address, acquiring the occupation situation of a fire fighting channel, and acquiring the contact way of people occupying the fire fighting channel according to the occupation situation of the fire fighting channel to carry out intelligent reminding;

when the contact information of the person occupying the fire fighting access cannot be acquired, executing alarm processing, and sending the fire fighting access occupation information to an alarm port;

when the method is implemented, the alarm information of the fire fighting team for giving an alarm and the corresponding accident reason and accident address are acquired, a mapping relation is established between the accident address and the accident reason, the accident is predicted based on a preset big data information platform, the predicted accident address is determined, corresponding monitoring video information and fire fighting channel distribution information in an accident address area are acquired, fire fighting channel occupation information is acquired, intelligent reminding is carried out aiming at the contact way of the person occupying the fire fighting channel, and when the contact way of the person occupying the fire fighting channel cannot be acquired, alarm processing is carried out;

the beneficial effects of the above technical scheme are: the accident address and the accident reason are mapped, the corresponding monitoring video information and the corresponding fire fighting channel distribution information in the accident address area are obtained after the accident is predicted based on a preset big data information platform, the fire fighting channel occupation information is obtained, the fire fighting channel occupation condition of an important road section is predicted and detected in advance, and the dangerous condition caused by the occupation of the fire fighting channel is reduced.

Example 10:

the embodiment of the invention provides a fire fighting access occupation recognition device based on an artificial intelligence technology, as shown in the attached figure 4, comprising:

an image collector: the system comprises a monitoring area, a fire fighting channel and a video processing unit, wherein the monitoring area is used for acquiring video information in a monitoring area corresponding to the fire fighting channel, and performing video frame segmentation according to the video information to acquire a plurality of frames of image data;

fire-fighting access occupation detector: the system is used for judging whether the fire fighting channel is occupied or not according to the image data based on a preset target detection system, and if so, determining an occupied object;

fire control passageway occupies the alarm: the device is used for determining corresponding occupied time according to the occupied object, judging whether the occupied time is greater than a preset alarm threshold value or not, and if so, executing alarm processing;

in one practical scenario: the method is only suitable for detecting continuous vehicle video frames, and the calculation amount of the algorithm is large and the corresponding time complexity is high in the method, so that the performance requirement on hardware equipment is strict and the relative maintenance cost is high;

when the method is implemented and aiming at the occupation condition detection of the fire fighting access, the occupation condition of the fire fighting access area is detected, the occupied object and the occupation time are determined, the calculation efficiency is high, the calculation complexity is low, the adaptability to the change of the illumination condition is strong, the detailed outline of the occupied object is not required to be extracted, only the edge outline is required to be extracted, and the result is preliminarily determined;

the beneficial effects of the above technical scheme are: in this technical scheme at first carry out the video frame according to the video frame data in the monitoring video region and cut apart, acquire a plurality of frame image data, but not select whole video frame data, be favorable to reducing the complexity of calculation, improve the operating efficiency, secondly, through discerning fire control passageway occupation object, be favorable to occupying different processing methods of intelligence selection according to the difference, make the processing to fire control passageway occupation have more pertinence, report to the police to the occupation of fire control passageway through the mode of presetting the time threshold value at last, be favorable to reducing the speed and the efficiency that fire control passageway was handled, effectively avoid the condition that fire control passageway occupied to take place.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (10)

1. A fire fighting access occupation identification method based on an artificial intelligence technology is characterized by comprising the following steps:

acquiring video information in a monitoring area corresponding to a fire fighting channel, and performing video frame segmentation according to the video information to acquire a plurality of frames of image data;

based on a preset target detection system, judging whether the fire fighting channel is occupied or not according to the image data, and if so, determining an occupied object;

and determining corresponding occupation time according to the occupation object, judging whether the occupation time is greater than a preset alarm threshold value, and if so, executing alarm processing.

2. The method for identifying the occupation of the fire fighting access based on the artificial intelligence technology as claimed in claim 1, wherein the step of obtaining the video information in the monitoring area corresponding to the fire fighting access, and performing video frame segmentation according to the video information to obtain a plurality of frames of image data comprises:

acquiring position information of a fire fighting access, judging whether a monitoring device exists in the fire fighting access area according to the position information, and determining a judgment result;

when the judgment result shows that the monitoring device exists in the fire fighting access area, acquiring video information of the monitoring area;

according to the video information of the monitoring area, carrying out area division on the video information, and determining an area division result; wherein the region division result comprises: fire-fighting vehicle driving areas, other areas;

and carrying out segmentation processing on the video information corresponding to the fire fighting vehicle driving area to obtain a plurality of frames of image data of the fire fighting vehicle driving area.

3. The method as claimed in claim 1, wherein the method for identifying fire fighting access occupation based on artificial intelligence technology, based on the preset target detection system, judges whether the fire fighting access is occupied according to the image data, and if yes, determines the occupied object, including:

performing image recognition on image data of a monitoring area corresponding to the fire fighting channel to obtain an image recognition result; wherein the image recognition result comprises: the fire fighting channel is occupied and the fire fighting channel is not occupied;

when the image recognition result shows that the driving area of the fire fighting vehicle is occupied, determining an occupied object occupying the driving area based on a preset target detection system; wherein the occupancy object comprises: automotive, non-automotive, object.

4. The method for identifying the occupation of the fire fighting access based on the artificial intelligence technology as claimed in claim 1, wherein the step of determining the corresponding occupation time according to the occupation object, the step of judging whether the occupation time is greater than a preset alarm threshold value, and if so, the step of executing alarm processing comprises the steps of:

setting a fire fighting channel time occupation threshold according to different occupied objects, and sending occupation information corresponding to a fire fighting channel to a preset video monitoring platform when the occupied time of the occupied object reaches a preset time occupation threshold according to the occupied time of the occupied object; wherein the occupancy information comprises: occupied objects, occupied time.

5. The method for identifying the occupation of the fire fighting access based on the artificial intelligence technology as claimed in claim 1, wherein the step of determining the corresponding occupation time according to the occupation object, determining whether the occupation time is greater than a preset alarm threshold, and if so, executing alarm processing further comprises:

when the occupied time of the occupied object is larger than a preset time occupation threshold, acquiring a target detection result of the occupied object;

when the occupied object is determined to be the motor vehicle, identifying and acquiring license plate information of the motor vehicle, acquiring vehicle owner information and a telephone number of the motor vehicle based on a preset big data information platform, and intelligently sending a reminding short message;

when the occupied object is a non-motor vehicle, acquiring an initial occupied time point of the occupied object according to monitoring video information, acquiring facial image data of a corresponding owner of the non-motor vehicle according to the initial occupied time point, extracting facial features according to the facial image data, constructing a facial information matrix according to the facial features, acquiring a matching result of a target object according to the facial information matrix based on the big data information platform, determining personnel information and a telephone number of the target object when the matching result is greater than a preset matching threshold value, and intelligently sending a reminding short message;

when the occupied object is an object, acquiring an initial occupied time point of the occupied object according to monitoring video information, acquiring face information of the owner corresponding to the object according to the initial occupied time point, acquiring personnel information and a telephone number of the owner of the object based on the big data information platform, and intelligently sending a reminding short message.

6. The method for identifying the occupation of the fire fighting access based on the artificial intelligence technology as claimed in claim 1, wherein the steps of obtaining video information in a monitoring area corresponding to the fire fighting access, performing video frame segmentation according to the video information, and obtaining a plurality of frames of image data further comprise: preprocessing the image data to obtain an image preprocessing result, wherein the preprocessing process comprises the following steps:

carrying out graying processing on the image data to obtain primary image data;

acquiring pixel point distribution corresponding to the primary image data, judging whether image noise exists in the primary image data according to the pixel point distribution result, and if so, performing noise reduction processing on the primary image data by adopting a spatial filtering method to acquire secondary image data; the spatial filtering method is used for calculating the gray value of a pixel point in image data through a template;

randomly sampling the secondary image data to obtain a signal discrete sample corresponding to the secondary image data, performing image reconstruction on the signal discrete sample by a nonlinear reconstruction method, and obtaining connected domain data information corresponding to the secondary image data according to an image reconstruction result; the nonlinear reconstruction method is used for obtaining low-frequency component loss of image data and performing high-frequency reconstruction aiming at the low-frequency component loss.

7. The method as claimed in claim 1, wherein the method for identifying the occupation of the fire fighting access based on the artificial intelligence technology is characterized in that the method for identifying the occupation of the fire fighting access based on the preset target detection system judges whether the fire fighting access is occupied according to the image data, and if yes, determines the occupation object, and further comprises:

acquiring fire station alarm information, determining the current location and target address of a fire truck according to the alarm information, planning a path according to the current location and target address of the fire truck, and determining a path planning result;

acquiring fire fighting channel setting information in the path according to the path planning result, acquiring monitoring information of a corresponding position according to the fire fighting channel setting information, judging whether a fire fighting channel is occupied or not according to the monitoring information, and determining a judgment result;

acquiring fire fighting channel occupation information when the judgment result shows that the situation that the fire fighting channel is occupied exists in the path; wherein the fire fighting access occupancy information comprises: the fire fighting access occupies the position and the object information;

and sending the fire fighting access occupation information to an emergency accident handling port.

8. The method for identifying the occupation of the fire fighting access based on the artificial intelligence technology as claimed in claim 1, wherein the step of obtaining the video information in the monitoring area corresponding to the fire fighting access, and performing video frame segmentation according to the video information to obtain a plurality of frames of image data comprises:

receiving fire fighting access occupation information uploaded by a mobile equipment port based on a preset big data information platform; the method for uploading information by the mobile equipment port comprises the following steps: wireless communication mode uploads, the phone reports and uploads, fire control passageway occupies information and includes: occupation address and occupation vehicle information;

performing data integration aiming at the fire fighting access occupation information to acquire sequence information occupied by the fire fighting access;

and acquiring vehicle license plate information occupying the fire fighting access according to the sequence information occupied by the fire fighting access, acquiring a contact way of a host according to the license plate information, and intelligently reminding according to the contact way.

9. The method for identifying the occupation of the fire fighting access based on the artificial intelligence technology as claimed in claim 1, wherein the steps of obtaining video information in a monitoring area corresponding to the fire fighting access, performing video frame segmentation according to the video information, and obtaining a plurality of frames of image data further comprise:

acquiring corresponding accident reasons and accident addresses according to alarm information of fire police, establishing a mapping relation between the accident addresses and the accident reasons, predicting the accident based on a preset big data information platform, and determining a predicted accident address;

acquiring corresponding monitoring video information and fire fighting channel distribution information in the accident address area according to the predicted accident address, acquiring the occupation situation of a fire fighting channel, and acquiring the contact way of people occupying the fire fighting channel according to the occupation situation of the fire fighting channel to carry out intelligent reminding;

and when the contact information of the person occupying the fire fighting access cannot be acquired, executing alarm processing, and sending the fire fighting access occupation information to an alarm port.

10. The fire fighting access occupancy recognition device based on artificial intelligence technology as recited in claim 1, comprising:

an image collector: the system comprises a monitoring area, a fire fighting channel and a video frame segmentation module, wherein the monitoring area is used for acquiring video information in a monitoring area corresponding to the fire fighting channel, and performing video frame segmentation according to the video information to acquire a plurality of frames of image data;

fire-fighting access occupation detector: the system is used for judging whether the fire fighting channel is occupied or not according to the image data based on a preset target detection system, and if so, determining an occupied object;

fire control passageway occupies the alarm: and the device is used for determining corresponding occupied time according to the occupied object, judging whether the occupied time is greater than a preset alarm threshold value or not, and executing alarm processing if the occupied time is greater than the preset alarm threshold value.

Images