CN113255588A - Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium - Google Patents

Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium Download PDF

Info

Publication number
CN113255588A
CN113255588A CN202110701888.1A CN202110701888A CN113255588A CN 113255588 A CN113255588 A CN 113255588A CN 202110701888 A CN202110701888 A CN 202110701888A CN 113255588 A CN113255588 A CN 113255588A
Authority
CN
China
Prior art keywords
target
garbage
frame
tracking
sweeper
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110701888.1A
Other languages
Chinese (zh)
Other versions
CN113255588B (en
Inventor
汪寒
何军强
顾鹏笠
陈戗
金丽娟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Hopechart Iot Technology Co ltd
Original Assignee
Hangzhou Hopechart Iot Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Hopechart Iot Technology Co ltd filed Critical Hangzhou Hopechart Iot Technology Co ltd
Priority to CN202110701888.1A priority Critical patent/CN113255588B/en
Publication of CN113255588A publication Critical patent/CN113255588A/en
Application granted granted Critical
Publication of CN113255588B publication Critical patent/CN113255588B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • E—FIXED CONSTRUCTIONS
    • E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
    • E01H—STREET CLEANING; CLEANING OF PERMANENT WAYS; CLEANING BEACHES; DISPERSING OR PREVENTING FOG IN GENERAL CLEANING STREET OR RAILWAY FURNITURE OR TUNNEL WALLS
    • E01H1/00—Removing undesirable matter from roads or like surfaces, with or without moistening of the surface
    • E—FIXED CONSTRUCTIONS
    • E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
    • E01H—STREET CLEANING; CLEANING OF PERMANENT WAYS; CLEANING BEACHES; DISPERSING OR PREVENTING FOG IN GENERAL CLEANING STREET OR RAILWAY FURNITURE OR TUNNEL WALLS
    • E01H1/00—Removing undesirable matter from roads or like surfaces, with or without moistening of the surface
    • E01H1/005—Mobile installations, particularly for upkeeping in situ road or railway furniture, for instance road barricades, traffic signs; Mobile installations particularly for upkeeping tunnel walls
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/48—Matching video sequences
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • 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
    • G06T2207/10048—Infrared image
    • 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
    • G06T2207/30248—Vehicle exterior or interior
    • G06T2207/30252—Vehicle exterior; Vicinity of vehicle
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Structural Engineering (AREA)
  • Civil Engineering (AREA)
  • Architecture (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Image Analysis (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention provides a garbage cleaning method and device of a garbage sweeper, electronic equipment and a storage medium, wherein the method comprises the following steps: inputting an image set to be processed into a target detection network so as to determine target information included in the image set to be processed; filtering the target information according to a target preset condition and a preset filtering condition; and performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after determining that the target garbage needing to be cleaned exists in the tracking result. The device is used for executing the method. The garbage cleaning method, the device, the electronic equipment and the storage medium of the garbage sweeper can accurately detect road garbage and street overflow garbage cans in real time, and evaluate the real-time cleaning effect of the road surface of the garbage sweeper after operation, so that the dependence of the existing garbage sweeper on operators is reduced.

Description

Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium
Technical Field
The invention relates to the technical field of digital image processing, in particular to a garbage sweeping method and device for a garbage sweeper, electronic equipment and a storage medium.
Background
In recent years, more and more cities have achieved mechanization of road surface cleaning work. The method is not only beneficial to the improvement of urban modernization images, but also reduces the labor intensity of sanitation workers and improves the safety, the economy and the operation efficiency in the operation process.
Most of large garbage sweeper on the road surface at present are only manually opened for cleaning, the garbage can on the street is manually observed to determine whether to dump or not, partial information is often missed due to negligence due to excessive dependence on manual work, and the garbage sweeper at the present stage is difficult to evaluate whether the cleaned road surface achieves the expected cleaning effect or not in real time.
Disclosure of Invention
The garbage cleaning method, the garbage cleaning device, the electronic equipment and the storage medium of the garbage sweeper provided by the invention are used for overcoming at least one problem in the prior art, can accurately detect road surface garbage such as road surface garbage and street overflow garbage cans in real time, and cleans the detected garbage by using the garbage sweeper, so that the dependence of the existing garbage sweeper on operators is reduced.
The invention provides a garbage cleaning method of a garbage sweeper, which comprises the following steps:
inputting an image set to be processed into a target detection network so as to determine target information included in the image set to be processed;
filtering the target information according to a target preset condition and a preset filtering condition;
performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after confirming that the target garbage needing to be cleaned exists according to the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot by a target camera;
the target cameras are respectively installed at different target positions of the garbage sweeper.
According to the garbage cleaning method of the garbage sweeper provided by the invention, the target information is filtered according to the target preset condition, and the method comprises the following steps:
if the target camera is installed at a first target position of the garbage sweeper, target information belonging to a first type of target garbage is screened out;
if the target camera is installed at a second target position of the garbage sweeper, target information belonging to a second type of target garbage is screened out;
if the target camera is installed at a third target position of the garbage sweeper, target information belonging to the first type of target garbage is screened out;
wherein the first target location is included within a front windshield;
the second target position comprises a primary-secondary outboard top edge;
the third target position comprises a vehicle rear middle position.
According to the garbage cleaning method of the garbage sweeper provided by the invention, the multiframe tracking is carried out on the filtering result based on the multiframe tracking algorithm, and the method comprises the following steps:
acquiring a tracking frame list and a detection frame list;
respectively determining the predicted positions meeting a first preset condition and a second preset condition;
performing bidirectional matching on the tracking frame list and the detection frame list according to a preset matching rule, and determining a first target tracking frame successfully subjected to bidirectional matching from the tracking frame list;
updating the number of times that the first target tracking frame is matched by the detection frame, the first position of the first target tracking frame, the second position of the first target tracking frame, the first mismatch frame number of the first target tracking frame and the second mismatch frame number of the first target tracking frame according to a first preset updating rule;
updating a first mismatch frame number of a second target tracking frame with bidirectional matching failure according to a second preset updating rule, and removing the second target tracking frame from the tracking frame list when the first mismatch frame number of the second target tracking frame is greater than a preset maximum continuous mismatch frequency so as to update the tracking frame list;
adding a detection box with a failure in bidirectional matching to the tracking box list so as to update the tracking box list;
traversing the updated tracking frame list, and taking the tracking frame which is matched by the detection frame in the updated tracking frame list for a time not less than a preset minimum matching time as the tracking result;
the tracking frame list is obtained according to the predicted position of a tracking frame, the first position of the last detection frame matched with the tracking frame, the second position of the target detection frame matched with the tracking frame, the first mismatch frame number of the predicted position and the first position, the second mismatch frame number of the first position and the second position, the matching times of the tracking frame by the detection frame and the target category of the tracking frame;
and the detection frame list is obtained according to the coordinates of the detection frame, the target category of the detection frame and the preset filtering condition.
According to the garbage cleaning method of the garbage sweeper provided by the invention, the step of respectively determining the predicted positions meeting the first preset condition and the second preset condition comprises the following steps:
if the first preset condition is met, determining the predicted position based on the first position, the first mismatch frame number and the offset;
if the second preset condition is met, determining the predicted position according to the first position;
wherein the first preset condition includes the presence of a tracking frame for the first location and the second location;
the second preset condition comprises that no tracking frame of the second position exists;
the offset is an offset of each frame coordinate predicted from the first location, the second location, and the second mismatch frame number.
According to the garbage cleaning method of the garbage sweeper provided by the invention, the preset matching rule comprises the following steps:
traversing the detection frame list, obtaining the intersection and parallel ratio of the predicted position and the first position, and determining a detection frame matched with the tracking frame according to a third preset condition;
taking the detection frame with the highest matching score with the tracking frame as the first target tracking frame;
the third preset condition comprises that the intersection ratio is larger than a detection frame with the largest intersection ratio in preset values.
According to the garbage cleaning method of the garbage sweeper provided by the invention, the preset filtering condition comprises the following steps:
and eliminating the detection frame with the confidence coefficient smaller than a first preset threshold value.
According to the garbage cleaning method of the garbage sweeper provided by the invention, if the image set to be processed is composed of road surface images shot and acquired by the target camera arranged at the third target position, the first type of target garbage cleaning operation is detected by the following method:
merging one or more adjacent tracking frames with the same target category in the tracking result based on a target rectangular frame clustering algorithm to obtain a clustered target frame;
and acquiring the coverage rate according to the ratio of the rectangular areas after the overlapping parts of all the target frames are removed, and determining that the first type of residual target garbage exists when the coverage rate is smaller than a preset threshold value.
The invention also provides a garbage cleaning device of the garbage sweeper, which comprises: the system comprises an image processing module, a data processing module and a garbage cleaning module;
the image processing module is used for inputting the image set to be processed into a target detection network so as to determine target information included in the image set to be processed;
the data processing module is used for filtering the target information according to a target preset condition and a preset filtering condition;
the garbage cleaning module is used for carrying out multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after determining that the target garbage needing to be cleaned exists in the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot by a target camera;
the target cameras are respectively installed at different target positions of the garbage sweeper.
The invention also provides an electronic device, which 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 garbage cleaning method of the garbage sweeper.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method for sweeping the garbage of a garbage sweeper as described in any one of the above.
The garbage cleaning method, the device, the electronic equipment and the storage medium of the garbage sweeper provided by the invention can accurately and reliably detect the garbage on the road surface such as the road surface and the street overflow garbage can, and clean the detected garbage by using the garbage sweeper, thereby reducing the dependence of the existing garbage sweeper on operators.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for sweeping garbage of a garbage sweeper according to the present invention;
FIG. 2 is a schematic view of the garbage sweeping system of the garbage sweeper provided by the present invention;
FIG. 3 is a schematic view of the installation position of the target camera provided by the present invention;
FIG. 4 is a schematic diagram of rectangular calculation of the rectangular box clustering algorithm provided by the present invention;
FIG. 5 is a schematic structural view of a garbage sweeping device of the garbage sweeper provided by the present invention;
fig. 6 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart of a method for sweeping garbage of a garbage sweeper according to the present invention, and as shown in fig. 1, the method includes:
s1, inputting the image set to be processed into a target detection network to determine target information included in the image set to be processed;
s2, filtering the target information according to a target preset condition and a preset filtering condition;
s3, performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after determining that the target garbage needing to be cleaned exists in the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot and acquired by a target camera;
the target cameras are respectively arranged at different target positions of the garbage sweeper.
It should be noted that the main body of the method may be a computer device, or may be a schematic structural diagram of the garbage sweeper shown in fig. 2. The following describes the garbage cleaning method of the garbage sweeper in detail by taking the garbage sweeper shown in fig. 2 as an example:
optionally, as shown in fig. 2, the garbage sweeper includes a target camera, a vehicle-mounted terminal, a multimedia central control screen, and a background server. The system comprises a garbage sweeper, target cameras, a vehicle-mounted terminal and a control system, wherein the target cameras are arranged in four directions, namely front, back, left and right, of the garbage sweeper, are mainly used for monitoring the surrounding situation of the garbage sweeper and sending acquired road images to the vehicle-mounted terminal for reasoning; the vehicle-mounted terminal infers an image set acquired by the target camera, provides road surface garbage distribution and overflow conditions of the street garbage can, and transmits information to the multimedia central control screen and the background server; the multimedia central control screen is used for receiving road surface garbage information transmitted by the vehicle-mounted terminal, superposing a result on a video picture acquired by the camera and displaying the result to an operator, and giving a voice prompt when the garbage bin overflows and the cleaning effect is poor; the background server is used for receiving the cleaning effect evaluation information transmitted by the vehicle-mounted terminal, is convenient to record and store, and is used for a manager to monitor the working state of the sweeper.
Specifically, a background server is placed in a machine room provided by a user, and a vehicle management system is deployed in the server; the vehicle-mounted terminal and the multimedia central control screen are arranged in the cab, and the mounting positions of the vehicle-mounted terminal and the multimedia central control screen do not have too strict requirements, so that the vehicle-mounted terminal and the multimedia central control screen can be conveniently operated and watched by a driver and can be mounted at a position which does not interfere with the normal driving sight of the driver; the target cameras are respectively installed at different target positions of the garbage sweeper and used for shooting road surface images in real time to obtain an image set to be processed.
And sending the acquired image set to be processed into an AI chip of the vehicle-mounted terminal, zooming the road surface images in the image set to be processed by using the AI chip, and inputting the zoomed road surface images into a target detection network to obtain target information included in each road surface image in the image set to be processed. Wherein the road surface image can be scaled to 300 × 3 (where 300 represents the length and width of the image, and 3 represents the image divided into three channels of RGB), the target detection network can adopt a Mobilenetv2 (SSD) detection network.
The target information of the road surface image specifically includes: target location, target class, confidence, and detection box. The target position is determined by the coordinates of the central point x and y and the length and width of the detection frame, and the target category comprises the following eight types of target garbage: the garbage can comprises a non-overflowing fixed garbage can, an overflowing fixed garbage can, a non-overflowing tourist street garbage can, an overflowing tourist street garbage can, white garbage, leaves, mud stains and household garbage. The confidence in the target information is the confidence (0-100) of the bounding box.
The target detection network not only needs to predict the bounding boxes (bounding boxes) of each target garbage in the road surface image, but also needs to give the classification probability of each target. Typically, object detection predicts many bounding boxes. Each bounding box also needs a confidence score (confidence score) representing how likely it contains an object.
And then filtering the target information of each image in the image set to be processed according to a target preset condition and a preset filtering condition, carrying out multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after determining that the target garbage to be cleaned exists in the tracking result.
The garbage cleaning method of the garbage sweeper provided by the invention can accurately detect the garbage on the road surface such as the road surface and the street overflow garbage bin in real time, and clean the detected garbage by using the garbage sweeper, thereby reducing the dependence of the existing garbage sweeper on operators.
Further, in an embodiment, the filtering the target information according to the target preset condition in step S2 may specifically include:
s21, screening out target information belonging to the first type of target garbage if the target camera is installed at a first target position of the garbage sweeper;
s22, screening out target information belonging to a second type of target garbage if the target camera is installed at a second target position of the garbage sweeper;
s23, screening out target information belonging to the first type of target garbage if the target camera is installed at a third target position of the garbage sweeper;
wherein the first target location is included within the front windshield;
the second target position comprises a primary-secondary outboard top edge;
the third target position includes a tailstock center position.
Optionally, as shown in fig. 3, the target camera includes 4 infrared cameras, which are respectively marked as a camera a, a camera B, a camera C, and a camera D, where the camera a is installed in a position not affected by the wiper in the front windshield of the vehicle, and occupies all the images on the road surface; the camera B is arranged on the edge of the top of the outer side of the main driving of the vehicle; the camera C is arranged on the edge of the top of the outer side of the copilot of the vehicle; the camera D is installed in the middle of the tail of the vehicle, and the road surface occupies all pictures. And filtering the target information of the road surface image shot by the target cameras positioned at different target positions of the garbage sweeper according to the preset target condition. Specifically, the method comprises the following steps:
and screening out the target information belonging to the first class of target garbage if the target information of the image set to be processed is shot and acquired by a camera A arranged at a first target position of the garbage sweeper. The first type of target garbage comprises white garbage, leaves, mud stains and household garbage, and the camera A is used for detecting whether garbage exists on the front road surface.
The confidence coefficient is smaller than a first preset threshold valuepFiltering the target information of (1)pConfigurable, default to 50) and then performs multi-frame tracking on the filter results to remove false boxes.
After the fact that the garbage needing to be cleaned is the first-class target garbage (white garbage, leaves, mud stains and household garbage) is confirmed through a multi-frame tracking algorithm, the vehicle-mounted terminal sends a cleaning operation starting signal, the target position and the target class of the cleaned garbage are transmitted to the multimedia central control screen, and the multimedia central control screen overlaps frames on a video picture and displays the frames to operators.
And if the first type of target garbage does not exist, the vehicle-mounted terminal sends a signal for stopping cleaning operation.
And if the target information of the image set to be processed is shot and acquired by the camera B and the camera C which are arranged at the second target position of the garbage sweeper, screening out the target information belonging to the second type of target garbage. The second type of target garbage comprises an overflowing fixed garbage can and an overflowing street garbage can, and the camera B and the camera C are mainly used for detecting whether the overflowing garbage can exists on two sides of the garbage sweeper or not.
The confidence coefficient is smaller than a first preset threshold valuepAnd filtering the target information, and performing multi-frame tracking on the filtering result by adopting a multi-frame tracking algorithm to remove error frames.
After the situation that second-class target garbage is found in the overflowing garbage bin is confirmed through a multi-frame tracking algorithm, the vehicle-mounted terminal sends the position and the category of the overflowing garbage bin to the multimedia central control screen, the multimedia central control screen enables frames to be superposed on video pictures sent back by corresponding cameras and shows the video pictures to operators, and the fact that the left side/the right side of the vehicle-mounted terminal has the garbage bin to be dumped is reported.
And screening out the target information belonging to the first class of target garbage if the target information of the image set to be processed is shot and acquired by a camera D arranged at a third target position of the garbage sweeper. Wherein, camera D is used for carrying out the aassessment to cleaning effect.
The garbage cleaning method of the garbage sweeper provided by the invention can accurately detect the road garbage and the street overflow garbage bin in real time, and can monitor the road surface of the garbage sweeper after operation in real time.
Further, in an embodiment, the performing multi-frame tracking on the filtering result based on the multi-frame tracking algorithm in the step S3 may specifically include:
s31, acquiring a tracking frame list and a detection frame list;
s32, respectively determining the predicted positions meeting the first preset condition and the second preset condition;
s33, performing bidirectional matching on the tracking frame list and the detection frame list according to a preset matching rule, and determining a first target tracking frame successfully matched in a bidirectional mode from the tracking frame list;
s34, updating the matching times of the first target tracking frame to the detected frame, the first position of the first target tracking frame, the second position of the first target tracking frame, the first mismatch frame number of the first target tracking frame and the second mismatch frame number of the first target tracking frame according to a first preset updating rule;
s35, updating the first mismatch frame number of the second target tracking frame with bidirectional matching failure according to a second preset updating rule, and removing the second target tracking frame from the tracking frame list when the first mismatch frame number of the second target tracking frame is greater than a preset maximum continuous mismatch frequency so as to update the tracking frame list;
s36, adding the detection box with the failure of the two-way matching into the tracking box list to update the tracking box list;
s37, traversing the updated tracking frame list, and taking the tracking frame, which is matched by the detected frame in the updated tracking frame list for a time not less than the preset minimum matching time, as a tracking result;
the tracking frame list is obtained according to the predicted position of the tracking frame, the first position of the last detection frame matched with the tracking frame, the second position of the target detection frame matched with the tracking frame, the first mismatch frame number of the predicted position and the first position, the second mismatch frame number of the first position and the second position, the matching times of the tracking frame to the detection frame and the target category of the tracking frame;
the detection frame list is obtained according to the coordinates of the detection frame, the target category to which the detection frame belongs and the preset filtering condition.
Further, in an embodiment, the presetting of the filtering condition may specifically include:
and eliminating the detection frames with the confidence degrees smaller than a first preset threshold value.
Optionally, a list ListT of trace boxes is determined, each trace box being defined as BoxT, each trace box recording the following information: the predicted position of the current frame of the tracking framelnFirst position of a matched detection frame on the tracking frameln-1, second position of last matched detection box (i.e. target detection box) on the tracking boxln-2，ln-1 andln-2 second mismatch frame numberfn-2，lnAndln-a first mismatch frame number between 1fn-1 number of times the tracking box is matched by the detection boxdThe target category to which the tracking box belongs.
Determining a list ListD of test boxes, each defined as BoxD, each recording the following information: coordinates of the detection framelAnd detecting the target category of the frame.
Determining a preset maximum number of consecutive mismatchesmAnd preset minimum matching timest。
After passing through the target detection network, the filtering confidence coefficient is lower than a first preset threshold valuepAfter the detection of the frame, a list ListD of the detection frame is obtained. And determining the predicted positions meeting the first preset condition and the second preset condition, performing bidirectional matching on the tracking frame list ListT and the detection frame list ListD according to a preset matching rule, and determining a first target tracking frame successfully subjected to bidirectional matching from the tracking frame list ListT.
For the first target tracking frame successfully matched in the two directions, the times of matching the first target tracking frame to the detected frame based on the first preset updating ruledA first position of the first target tracking frameln-1, second position of first target tracking frameln-2, first mismatch frame number of first target tracking framefn-1 and a second mismatch frame number for the first target tracking framefn-2 updating:d=d+1，ln-2=ln-1，ln-1=l，fn-2=fn-1，fn-1=0。
for a second target tracking frame with bidirectional matching failure, based on a second preset updating rule, the number of first mismatch frames of the second target tracking framefn-1 is updated and willfn-1=fn-1+1 updated first mismatch frame number of second target tracking frame, if updatedfn-1>mRemoving the second target tracking box from the tracking box list ListT to update the tracking box list ListT;
adding the detection box with the failure of bidirectional matching into a tracking box list ListT to update the tracking box list ListT;
traversing the updated tracking frame list and matching the times of the detected frames in the updated tracking frame listdNot less than a preset minimum number of matchestAs a result of the tracking.
The garbage cleaning method of the garbage sweeper provided by the invention can improve the target tracking capability when the image set to be processed is shielded by multiple frames, and further improve the recognition capability of the road garbage.
Further, in an embodiment, the step S32 may specifically include:
s321, if a first preset condition is met, determining a predicted position based on the first position, the first mismatch frame number and the offset;
s322, if a second preset condition is met, determining a predicted position according to the first position;
the first preset condition comprises that a tracking frame of a first position and a second position exists;
the second preset condition comprises that no tracking frame of the second position exists;
the offset is an offset of coordinates of each frame predicted from the first position, the second position, and the second mismatch frame number.
Alternatively, if the first preset condition is satisfied, that is, for the presenceln-1，ln-2 tracking box boxT, withln-1，ln-2，fn-2 the offset of each frame coordinate can be predictedvThen pass throughln-1，fn-1，vTo deduce the predicted positionln(ii) a If the second predetermined condition is satisfied, i.e. for absenceln2, the trace box BoxT newly added to the list of trace boxes ListT from the list of test boxes ListD in the previous round of trace, is used directlyln-1 as predicted positionln。
According to the garbage cleaning method of the garbage sweeper, the predicted position of the tracking frame is obtained, and a foundation is laid for bidirectional matching of the tracking frame list and the detection frame list based on the predicted position.
Further, in an embodiment, the preset matching rule in step S33 may specifically include:
s331, traversing the detection frame list, acquiring the intersection ratio of the predicted position and the first position, and determining a detection frame matched with the tracking frame according to a third preset condition;
s332, taking the detection frame with the highest matching score with the tracking frame as a first target tracking frame;
the third preset condition comprises that the intersection ratio is larger than a detection frame with the largest intersection ratio in the preset values.
Optionally, performing bidirectional matching on targets in the detection box list ListD and the tracking box list ListT according to a preset matching rule, where the matching rule is: aiming at the tracking box boxT in the tracking box list, traversing the detection box list D, and calculating the predicted position of the corresponding tracking box boxTlnAnd coordinates of a detection box BoxD in the detection box listlAccording to a third preset condition, selecting the maximum value of iou in the detection frame BoxD with the iou larger than 0.1 as a matching object of the tracking frame BoxT, and updating the first mismatch frame number with the iou and the corresponding tracking frame BoxTfn-a quotient of 1+1 as the match score of the detection box BoxD reverse match tracking box BoxT.
After performing the one-way matching of the trace box BoxT and the detection box BoxD, each trace box BoxT may be matched to 0 or 1 detection box BoxD, and the detection box BoxD may be matched to the trace box BoxT equal to or greater than 0, by selecting the trace box BoxT with the highest matching score recorded in the detection box BoxD as the final match (i.e., the first target trace box).
According to the garbage cleaning method of the garbage sweeper, provided by the invention, the accuracy and reliability of the pavement garbage identification are improved through the bidirectional matching of the detection frame and the tracking frame.
Further, in an embodiment, after step S3, the method may further include:
if the image set to be processed is composed of road surface images shot and acquired by a target camera arranged at a third target position, the first type of target garbage cleaning operation is detected in the following mode:
merging one or more adjacent tracking frames with the same target category in the tracking result based on a target rectangular frame clustering algorithm to obtain a clustered target frame;
and acquiring the coverage rate according to the ratio of the rectangular areas after the overlapping parts of all the target frames are removed, and determining that the first type of residual target garbage exists when the coverage rate is smaller than a preset threshold value.
Optionally, when the image set to be processed is shot and acquired by the camera D, the first type of target garbage such as white garbage, leaves, mud and domestic garbage is screened out for the target information, and the confidence is lower than a preset thresholdpAnd then carrying out multi-frame tracking on the filtering result by adopting a multi-frame tracking algorithm so as to remove error frames.
And performing density-based rectangular frame clustering on the tracking result based on a target rectangular frame clustering algorithm, and taking a plurality of adjacent tracking frames with the same category as the same target frame to unify judgment criteria. When the garbage piles appear in the picture, the target result output through the deep learning may have the condition that a part of targets are separately selected in a frame mode and a part of targets are combined in a frame mode, and the difference exists in the coverage rate of the garbage piles.
And removing overlapped parts in all the target frames obtained after clustering, calculating the ratio of the area covered by the rectangle, and finally taking the coverage percentage of 100-percent as the score of cleaning operation.
After receiving the cleaning operation score transmitted by the vehicle-mounted terminal, the multimedia center control screen superposes the cleaning operation score on a video picture corresponding to the camera D, and when the score is smaller than a second preset threshold valuesAnd broadcasting 'residual garbage exists' in time, and uploading the coordinates and scores of the current position to a background server. Wherein the second preset threshold valuesIt can be configured, default to 80.
Target rectangular box clustering algorithm: the algorithm is modified from a density-based clustering algorithm DBSCAn, the point with the minimum unit in the original algorithm is replaced by a rectangle, and the distance between the point and the point is defined as the distance between two rectangles.
All rectangles in the target rectangular frame clustering algorithm are parallel to the coordinate axis, so that only 4 relations exist between every two rectangles, and the corresponding distances are defined as follows, and detailed description is given in fig. 4:
a. the two rectangles are overlapped on the x axis and the y axis, namely the two rectangles have an overlapping area, and the distance is 0;
b. the two rectangles do not intersect, the two rectangles are partially overlapped in the x-axis direction, and the minimum distance is the distance between the lower edge line of the upper rectangle and the upper edge line of the lower rectangle;
c. the two rectangles do not intersect, the two rectangles are partially overlapped in the y-axis direction, and the minimum distance is the distance between the right side line of the left rectangle and the left side line of the right rectangle;
d. the two rectangles do not intersect, do not coincide in both the x-axis and y-axis directions, and the minimum distance is the distance between the two vertices that are closest.
The main objective of the target rectangular box clustering algorithm is to merge adjacent tracking boxes, so the parameter minimum inclusion point matching value of DBScan is set to 1 here.
The garbage cleaning method of the garbage sweeper can accurately detect the road garbage and the street overflow garbage can in real time, and can evaluate the real-time cleaning effect of the road surface of the garbage sweeper after operation.
The following describes the garbage cleaning device of the garbage sweeper provided by the invention, and the garbage cleaning device of the garbage sweeper described below and the garbage cleaning method of the garbage sweeper described above can be referred to correspondingly.
Fig. 5 is a schematic structural view of a garbage cleaning device of a garbage sweeper according to the present invention, as shown in fig. 5, including: an image processing module 510, a data processing module 511 and a garbage cleaning module 512;
an image processing module 510, configured to input the image set to be processed into a target detection network, so as to determine target information included in the image set to be processed;
the data processing module 511 is configured to filter target information according to a target preset condition and a preset filtering condition;
the garbage cleaning module 512 is used for performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage cleaning vehicle after determining that the target garbage needing to be cleaned exists in the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot and acquired by a target camera;
the target cameras are respectively arranged at different target positions of the garbage sweeper.
The garbage cleaning device of the garbage sweeper provided by the invention can accurately and reliably detect garbage on roads such as road surfaces and street overflow garbage cans, and clean the detected garbage by using the garbage sweeper, so that the dependence of the existing garbage sweeper on operators is reduced.
Fig. 6 is a schematic structural diagram of an electronic device provided in the present invention, and as shown in fig. 6, the electronic device may include: a processor (processor) 610, a communication interface (communication interface) 611, a memory (memory) 612 and a bus (bus) 613, wherein the processor 610, the communication interface 611 and the memory 612 communicate with each other via the bus 613. The processor 610 may call logic instructions in the memory 612 to perform the following method:
inputting the image set to be processed into a target detection network so as to determine target information included in the image set to be processed;
filtering target information according to a target preset condition and a preset filtering condition;
performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after determining that the target garbage needing to be cleaned exists in the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot and acquired by a target camera;
the target cameras are respectively arranged at different target positions of the garbage sweeper.
In addition, the logic instructions in the memory may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention or a part thereof, which essentially contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer power screen (which may be a personal computer, a server, or a network power screen, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like.
Further, the present invention discloses a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions, which when executed by a computer, the computer is capable of executing the garbage cleaning method of the garbage sweeper provided by the above-mentioned method embodiments, for example, the method comprises:
inputting the image set to be processed into a target detection network so as to determine target information included in the image set to be processed;
filtering target information according to a target preset condition and a preset filtering condition;
performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after confirming that the target garbage needing to be cleaned exists according to the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot and acquired by a target camera;
the target cameras are respectively arranged at different target positions of the garbage sweeper.
In another aspect, the present invention also provides a non-transitory computer readable storage medium, on which a computer program is stored, the computer program being implemented by a processor to execute the garbage cleaning method of the garbage sweeper provided in the above embodiments, for example, the method includes:
inputting the image set to be processed into a target detection network so as to determine target information included in the image set to be processed;
filtering target information according to a target preset condition and a preset filtering condition;
performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after confirming that the target garbage needing to be cleaned exists according to the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot and acquired by a target camera;
the target cameras are respectively arranged at different target positions of the garbage sweeper.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the above technical solutions may be essentially or partially implemented in the form of software products, which may be stored in computer readable storage media, such as ROM/RAM, magnetic disk, optical disk, etc., and include instructions for causing a computer power supply screen (which may be a personal computer, a server, or a network power supply screen, etc.) to execute the method according to the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A garbage cleaning method of a garbage sweeper is characterized by comprising the following steps:
inputting an image set to be processed into a target detection network so as to determine target information included in the image set to be processed;
filtering the target information according to a target preset condition and a preset filtering condition;
performing multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after determining that the target garbage needing to be cleaned exists in the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot by a target camera;
the target cameras are respectively installed at different target positions of the garbage sweeper.
2. The method for sweeping garbage of a garbage sweeper according to claim 1, wherein filtering the target information according to the target preset condition comprises:
if the target camera is installed at a first target position of the garbage sweeper, target information belonging to a first type of target garbage is screened out;
if the target camera is installed at a second target position of the garbage sweeper, target information belonging to a second type of target garbage is screened out;
if the target camera is installed at a third target position of the garbage sweeper, target information belonging to the first type of target garbage is screened out;
wherein the first target location is included within a front windshield;
the second target position comprises a primary-secondary outboard top edge;
the third target position comprises a vehicle rear middle position.
3. The method for sweeping garbage of a garbage sweeper according to claim 1, wherein the multi-frame tracking of the filtering result based on the multi-frame tracking algorithm comprises:
acquiring a tracking frame list and a detection frame list;
respectively determining the predicted positions meeting a first preset condition and a second preset condition;
performing bidirectional matching on the tracking frame list and the detection frame list according to a preset matching rule, and determining a first target tracking frame successfully subjected to bidirectional matching from the tracking frame list;
updating the number of times that the first target tracking frame is matched by the detection frame, the first position of the first target tracking frame, the second position of the first target tracking frame, the first mismatch frame number of the first target tracking frame and the second mismatch frame number of the first target tracking frame according to a first preset updating rule;
updating a first mismatch frame number of a second target tracking frame with bidirectional matching failure according to a second preset updating rule, and removing the second target tracking frame from the tracking frame list when the first mismatch frame number of the second target tracking frame is greater than a preset maximum continuous mismatch frequency so as to update the tracking frame list;
adding a detection box with a failure in bidirectional matching to the tracking box list so as to update the tracking box list;
traversing the updated tracking frame list, and taking the tracking frame which is matched by the detection frame in the updated tracking frame list for a time not less than a preset minimum matching time as the tracking result;
the tracking frame list is obtained according to the predicted position of a tracking frame, the first position of the last detection frame matched with the tracking frame, the second position of the target detection frame matched with the tracking frame, the first mismatch frame number of the predicted position and the first position, the second mismatch frame number of the first position and the second position, the matching times of the tracking frame by the detection frame and the target category of the tracking frame;
and the detection frame list is obtained according to the coordinates of the detection frame, the target category of the detection frame and the preset filtering condition.
4. A method for sweeping garbage of a garbage sweeper according to claim 3, wherein said determining the predicted positions satisfying a first preset condition and a second preset condition respectively comprises:
if the first preset condition is met, determining the predicted position based on the first position, the first mismatch frame number and the offset;
if the second preset condition is met, determining the predicted position according to the first position;
wherein the first preset condition includes the presence of a tracking frame for the first location and the second location;
the second preset condition comprises that no tracking frame of the second position exists;
the offset is an offset of each frame coordinate predicted from the first location, the second location, and the second mismatch frame number.
5. The method of claim 3, wherein the predetermined matching rules comprise:
traversing the detection frame list, obtaining the intersection and parallel ratio of the predicted position and the first position, and determining a detection frame matched with the tracking frame according to a third preset condition;
taking the detection frame with the highest matching score with the tracking frame as the first target tracking frame;
the third preset condition comprises that the intersection ratio is larger than a detection frame with the largest intersection ratio in preset values.
6. A method as claimed in any one of claims 1 or 3, wherein the predetermined filtering condition comprises:
and eliminating the detection frame with the confidence coefficient smaller than a first preset threshold value.
7. The method for cleaning garbage of a garbage sweeper according to claim 2, wherein if the image set to be processed is composed of road images captured by the target camera installed at the third target position, the first type of target garbage cleaning operation is detected by:
merging one or more adjacent tracking frames with the same target category in the tracking result based on a target rectangular frame clustering algorithm to obtain a clustered target frame;
and acquiring the coverage rate according to the ratio of the rectangular areas after the overlapping parts of all the target frames are removed, and determining that the first type of residual target garbage exists when the coverage rate is smaller than a second preset threshold value.
8. The utility model provides a rubbish motor sweeper rubbish cleaning device which characterized in that includes: the system comprises an image processing module, a data processing module and a garbage cleaning module;
the image processing module is used for inputting the image set to be processed into a target detection network so as to determine target information included in the image set to be processed;
the data processing module is used for filtering the target information according to a target preset condition and a preset filtering condition;
the garbage cleaning module is used for carrying out multi-frame tracking on the filtering result based on a multi-frame tracking algorithm, and cleaning the target garbage by using a garbage sweeper after determining that the target garbage needing to be cleaned exists in the tracking result;
the target information comprises a target position, a target category, a confidence coefficient and a detection frame;
the image set to be processed is composed of road surface images shot by a target camera;
the target cameras are respectively installed at different target positions of the garbage sweeper.
9. An electronic device 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 for cleaning garbage of the garbage sweeper according to any one of claims 1-7 when executing the computer program.
10. A non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program when executed by a processor implements the steps of the method for sweeping the garbage of the garbage sweeper according to any one of claims 1 to 7.
CN202110701888.1A 2021-06-24 2021-06-24 Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium Active CN113255588B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110701888.1A CN113255588B (en) 2021-06-24 2021-06-24 Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110701888.1A CN113255588B (en) 2021-06-24 2021-06-24 Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN113255588A true CN113255588A (en) 2021-08-13
CN113255588B CN113255588B (en) 2021-10-01

Family

ID=77189407

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110701888.1A Active CN113255588B (en) 2021-06-24 2021-06-24 Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN113255588B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114283387A (en) * 2022-03-08 2022-04-05 深圳市万物云科技有限公司 Intelligent garbage point cleaning work order generation method and device and related medium
WO2023050637A1 (en) * 2021-09-30 2023-04-06 上海仙途智能科技有限公司 Garbage detection
CN117575543A (en) * 2024-01-15 2024-02-20 西安卓越软件开发有限公司 Intelligent property management method and device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110853076A (en) * 2019-11-08 2020-02-28 重庆市亿飞智联科技有限公司 Target tracking method, device, equipment and storage medium
CN111429478A (en) * 2020-04-13 2020-07-17 展讯通信（上海）有限公司 Target tracking method and related equipment
KR20200113694A (en) * 2019-03-26 2020-10-07 (주) 오소 Very short-term real-time job sharing system and a method thereof

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20200113694A (en) * 2019-03-26 2020-10-07 (주) 오소 Very short-term real-time job sharing system and a method thereof
CN110853076A (en) * 2019-11-08 2020-02-28 重庆市亿飞智联科技有限公司 Target tracking method, device, equipment and storage medium
CN111429478A (en) * 2020-04-13 2020-07-17 展讯通信（上海）有限公司 Target tracking method and related equipment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
搜青岛: "每小时清扫上千平方米 智能清洁机器人金沙滩啤酒城内"开跑"", 《搜狐：HTTPS://WWW.SOHU.COM/A/412594834_100119403》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023050637A1 (en) * 2021-09-30 2023-04-06 上海仙途智能科技有限公司 Garbage detection
CN114283387A (en) * 2022-03-08 2022-04-05 深圳市万物云科技有限公司 Intelligent garbage point cleaning work order generation method and device and related medium
CN114283387B (en) * 2022-03-08 2022-06-24 深圳市万物云科技有限公司 Intelligent garbage point cleaning work order generation method and device and related medium
CN117575543A (en) * 2024-01-15 2024-02-20 西安卓越软件开发有限公司 Intelligent property management method and device
CN117575543B (en) * 2024-01-15 2024-04-30 西安卓越软件开发有限公司 Intelligent property management method and device

Also Published As

Publication number Publication date
CN113255588B (en) 2021-10-01

Similar Documents

Publication Publication Date Title
CN113255588B (en) Garbage cleaning method and device for garbage sweeper, electronic equipment and storage medium
CN112883819B (en) Multi-target tracking method, device, system and computer readable storage medium
CN102348128B (en) Surveillance camera system having camera malfunction detection function
CN111629181B (en) Fire-fighting life passage monitoring system and method
CN105741559B (en) A kind of illegal occupancy Emergency Vehicle Lane detection method based on track line model
CN112777169B (en) Internet of things monitoring method and system applied to garbage classification putting points
CN110659546B (en) Illegal booth detection method and device
CN107392139A (en) A kind of method for detecting lane lines and terminal device based on Hough transformation
JP2020170314A (en) Road abnormality detection system, road abnormality detection method, and road abnormality detection program
CN112818834B (en) Method, device and medium for judging avoidance of emergency vehicle at intersection
CN111967396A (en) Processing method, device and equipment for obstacle detection and storage medium
CN106910339A (en) Road information provides method, device and processing terminal
CN114170569A (en) Method, system, storage medium and equipment for monitoring road surface abnormal condition
CN115761668A (en) Camera stain recognition method and device, vehicle and storage medium
CN117173415A (en) Visual analysis method and system for large-scale floor washing machine
CN111105135A (en) Intelligent city sweeper operation monitoring method and device
KR101453386B1 (en) Vehicle Intelligent Search System and Operating Method thereof
CN115797880A (en) Method and device for determining driving behavior, storage medium and electronic device
CN114549444A (en) Cross-camera target detection and tracking system and method
CN115601972A (en) Obstacle processing system for unmanned mine driving area
CN114972731A (en) Traffic light detection and identification method and device, moving tool and storage medium
CN113902987A (en) Object detection method for group intelligence, storage medium and processor
CN112419778A (en) Parking lot vacancy detection and optimal path guiding device based on image processing and vacancy detection and optimal path guiding method thereof
CN116156307A (en) Camera cleaning method, device and equipment
CN106022336A (en) License plate detection method based on expressway snapshot image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant