CN114202700A - Cargo volume anomaly detection method and device and storage medium - Google Patents

Cargo volume anomaly detection method and device and storage medium Download PDF

Info

Publication number
CN114202700A
CN114202700A CN202111540672.8A CN202111540672A CN114202700A CN 114202700 A CN114202700 A CN 114202700A CN 202111540672 A CN202111540672 A CN 202111540672A CN 114202700 A CN114202700 A CN 114202700A
Authority
CN
China
Prior art keywords
cargo
goods
image
track
volume
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
CN202111540672.8A
Other languages
Chinese (zh)
Other versions
CN114202700B (en
Inventor
梁帆
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.)
Guangdong Prophet Big Data Co ltd
Original Assignee
Dongguan Prophet Big Data 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 Dongguan Prophet Big Data Co ltd filed Critical Dongguan Prophet Big Data Co ltd
Priority to CN202111540672.8A priority Critical patent/CN114202700B/en
Publication of CN114202700A publication Critical patent/CN114202700A/en
Application granted granted Critical
Publication of CN114202700B publication Critical patent/CN114202700B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a cargo volume anomaly detection method and device and a storage medium, which are used for improving the efficiency of cargo volume anomaly detection. The application discloses cargo volume anomaly detection method includes: acquiring a second image of a transport tool in a refrigeration house, and adding a label to the second image according to the volume of the transport tool and the offset of the transport tool relative to a door of the refrigeration house to obtain a third image; training the third image to obtain the transport tool detection model; acquiring a fourth image of goods in a refrigeration house, and adding a label to the fourth image according to the type of a goods box to obtain a fifth image; training the fifth image to obtain a cargo detection model; collecting a sixth image of the refrigeration house, and determining a characteristic data set of a freight transportation track according to the sixth image, the transport tool detection model and the cargo detection model; and detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track. The application also provides a cargo volume anomaly detection device and a storage medium.

Description

Cargo volume anomaly detection method and device and storage medium
Technical Field
The present application relates to the field of computing technologies, and in particular, to a method and an apparatus for detecting an abnormal cargo volume, and a storage medium.
Background
Through comparing and checking the volume of the actual transported goods and the volume registered in the book, the situation that all goods can be added in the transportation process or the goods strictly forbidden by the country are prevented from happening, the supervision responsibility is implemented, and the supervision efficiency is enhanced. However, in the prior art, the automatic detection of the abnormal volume of the goods can not be realized without affecting the goods carrying, and the efficiency is low.
Disclosure of Invention
In view of the above technical problems, embodiments of the present application provide a method and an apparatus for detecting an abnormal cargo volume, and a storage medium, so as to detect whether the cargo volume is abnormal, and improve the efficiency of detecting the abnormal cargo volume.
In a first aspect, a cargo volume anomaly detection method provided in an embodiment of the present application includes:
acquiring a second image of a transport tool in a refrigeration house, and adding a label to the second image according to the volume of the transport tool and the offset of the transport tool relative to a door of the refrigeration house to obtain a third image;
training the third image to obtain the transport tool detection model;
acquiring a fourth image of goods in a refrigeration house, and adding a label to the fourth image according to the type of a goods box to obtain a fifth image;
training the fifth image to obtain a cargo detection model;
collecting a sixth image of the refrigeration house, and determining a characteristic data set of a freight transportation track according to the sixth image, the transport tool detection model and the cargo detection model;
and detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track.
Preferably, the determining a characteristic data set of a freight transportation trajectory according to the sixth image, the vehicle detection model and the cargo detection model includes:
detecting vehicle information in the sixth image using the vehicle detection model;
if the transport means exists in the sixth image, detecting the cargo information in the sixth image by using the cargo detection model;
determining a vehicle characteristic vector set according to the vehicle information and the cargo information;
determining a difference matching score of the transport means according to a transport means feature vector set obtained from the sixth images of the front frame and the rear frame;
determining a set of continuous trajectory feature vectors of the same transport means according to the difference matching score of the transport means;
and determining a characteristic data set of the incoming transportation track according to the transportation tool characteristic vector set and the continuous track characteristic vector set.
Preferably, the set of vehicle feature vectors is:
Figure BDA0003414198990000021
wherein,
xmis the horizontal axis coordinate of the transportation tool identification box center;
ymis the longitudinal axis coordinate of the vehicle identification box center;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
dstatusis the transport cargo state of the transport;
xcpthe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis the height of the vehicle identification box;
i is the transport number;
wherein the transport cargo state d of the transport meansstatusDetermined according to the following formula:
Figure BDA0003414198990000031
wherein, tsdtDistance threshold, ts, obtained for historical data trainingagTraining the obtained angle threshold value for historical data;
Figure BDA0003414198990000032
Figure BDA0003414198990000033
wherein A ═ 2xcp+wcp-2xgoods-wgoods，
B＝2ycp+hcp-2ygoods-hgoods,
θs2The deviation angle of the transportation tool relative to the straight line of the cold storage platform door is detected,
ε is a non-negative correction constant and satisfies ε < 1.
Preferably, the determining a set of continuous trajectory feature vectors of the same vehicle according to the difference matching score of the vehicle includes:
determining images of the same transport means according to the difference matching score;
combining the feature vectors of the images of the same transport means to obtain a continuous track feature vector set;
said determining images of the same vehicle from the discrepancy match scores comprises:
vehicle feature vector from previous frame
Figure BDA0003414198990000034
And the vehicle feature vector of the next frame
Figure BDA0003414198990000035
Calculating a disparity match score dij：
Figure BDA0003414198990000036
Wherein, cd1，cd2，cd3The weighting coefficient is obtained by training according to historical data;
when d isij＜tsdfDetermining that the transport means of the front frame and the rear frame are the same;
wherein, tsdfIs a preset threshold value.
Preferably, the determining the feature data set of the incoming and outgoing transportation trajectory according to the vehicle feature vector set and the continuous trajectory feature vector set includes:
calculating the freight state score of each continuous track, and judging that the continuous track is a freight track or an air freight track according to the freight state score;
extracting freight track data in the freight track, and judging whether the track is a freight input transport track or a freight output transport track;
extracting characteristic data of the goods input track to obtain a characteristic data set of the goods input and output track;
wherein the shipment status score is calculated according to the following formula:
Figure BDA0003414198990000041
wherein n isctThe number of elements in the feature set;
if d isct＞tsctJudging the freight track, otherwise, judging the freight track is the air freight track.
Preferably, the extracting the freight track data in the freight track and the judging that the track is the incoming transport track or the outgoing transport track includes:
for the ith feature vector
Figure BDA0003414198990000042
Calculated according to the following formula
Figure BDA0003414198990000043
And
Figure BDA0003414198990000044
equation 1:
Figure BDA0003414198990000045
equation 2:
Figure BDA0003414198990000046
wherein x isdIs the coordinate of the horizontal axis of the endpoint at the upper left corner of the cold storage door frame, ydIs the coordinate of the longitudinal axis of the endpoint at the upper left corner of the cold storage door frame, wdWidth of the door frame of the cold storagedIs the height, x of the door frame of the cold storagewIs the width of the cold storage screen, yhThe picture of the cold storage is high;
determining the characteristic vector of the transportation tool at the beginning of the track according to the formula 1 and the formula 2
Figure BDA0003414198990000047
Is/are as follows
Figure BDA0003414198990000048
And
Figure BDA0003414198990000049
determining a feature vector of a vehicle at the end of a trajectory according to said equations 1 and 2
Figure BDA00034141989900000410
Is/are as follows
Figure BDA00034141989900000411
And
Figure BDA00034141989900000412
if it is
Figure BDA00034141989900000413
And is
Figure BDA00034141989900000414
Judging as an incoming and outgoing freight track;
if it is
Figure BDA0003414198990000051
And is
Figure BDA0003414198990000052
Judging as an incoming and outgoing freight track;
wherein, tsbgAnd tsdgTraining the obtained threshold value for the historical data, and satisfying the following conditions:
Figure BDA0003414198990000053
where min () represents the minimum value.
Preferably, the detecting whether the cargo volume is abnormal according to the characteristic data set of the cargo transportation track includes:
estimating the volume V of goods transported on each goods-feeding track according to the characteristic data set of the goods-transporting trackiI is the number of the feeding track;
determining a total volume estimate for the cargo according to the following formula:
Vg＝∑Vi；
determining the volume V' of the goods and the quantity N of the goods according to the registration information of the goods, and calculating a registration value of the volume of the goods:
Vr＝V’N；
calculating a volume difference ratio P based on said estimated total volume and said registered volumed：
Figure BDA0003414198990000054
If b isd≤Pd≤buIf the volume of the goods is normal, otherwise, the volume of the goods is abnormal;
wherein, bdAnd buThe volume lower bound threshold and the volume upper bound threshold are obtained by training according to historical data.
Preferably, the estimation of the volume V of the goods transported in each goods-in track according to the characteristic data set of the goods-transporting trackiThe method comprises the following steps:
in the feature data set, the feature data is (x)cp,ycp,wcp,hcp,xgoods,ygoods,wgoods,hgoods,θ)；
When in use
Figure BDA0003414198990000055
The estimated cargo volume is:
Vi＝hcwclcPwlPh；
when in use
Figure BDA0003414198990000056
The estimated cargo volume is:
Figure BDA0003414198990000061
wherein x iscpThe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis the height of the vehicle identification box;
xgoodsis the horizontal axis coordinate of the endpoint of the upper left corner of the goods square frame;
ygoodsis the upper left corner end of the goods square frameCoordinates of a point longitudinal axis;
wgoodsis the width of the cargo box;
hgoodsis the height of the cargo box;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
hcis the vehicle height;
wcis the vehicle cross-sectional width;
lcis the tool length;
Pwlis an estimate of the floor zoom ratio of the good;
Phis an estimate of the cargo's height scaling ratio;
cvis a correction constant obtained by historical data training;
c0is the correction constant.
Further, the bottom surface scaling ratio estimate P of the goodwlDetermined by the following equation:
Figure BDA0003414198990000062
wherein, cwwIs a cross section scaling constant obtained by historical data training.
Further, an estimate of the cargo's height scaling ratio PhDetermined by the following equation:
Figure BDA0003414198990000063
wherein, chlIs the length scaling of the relative heights obtained by historical data training, chwWidth scaling of the relative height.
By using the cargo volume anomaly detection method provided by the invention, firstly, the image of the transport tool is collected and trained to obtain a detection model of the transport tool; then, collecting images of the goods, and training to obtain a goods detection model; and then acquiring an image of the refrigeration house, and detecting whether the cargo volume is abnormal or not according to the transport tool detection model and the cargo detection model. The cargo volume abnormity detection method provided by the invention can effectively detect the transported cargo volume in real time, compares the detected cargo volume with the registration information, and detects whether the cargo volume is abnormal.
In a second aspect, an embodiment of the present application further provides a cargo volume abnormality detection apparatus, including:
the data acquisition module is configured for acquiring a second image of the transport tool in the refrigeration house, a fourth image of the goods and a sixth image of the refrigeration house;
a cargo volume anomaly detection module configured to tag the second image according to the volume of the transport and the offset of the transport relative to the cold store door to obtain a third image; training the third image to obtain the transport tool detection model; adding a label to the fourth image according to the type of the cargo box to obtain a fifth image; training the fifth image to obtain a cargo detection model; determining a characteristic data set of a freight transportation track according to the sixth image, the transport detection model and the cargo detection model; and detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track.
In a third aspect, an embodiment of the present application further provides a cargo volume abnormality detection apparatus, including: a memory, a processor, and a user interface;
the memory for storing a computer program;
the user interface is used for realizing interaction with a user;
the processor is used for reading the computer program in the memory, and when the processor executes the computer program, the cargo volume abnormity detection method provided by the invention is realized.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a schematic flow chart of a cargo volume anomaly detection method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a feature data set for determining a transportation trajectory according to an embodiment of the present application;
fig. 3 is a schematic view of a cargo volume abnormality detection apparatus according to an embodiment of the present application;
fig. 4 is a schematic view of another cargo volume abnormality detection apparatus provided in the embodiment of the present application;
fig. 5 is a schematic view of an offset angle of a straight line where the refrigeration house door is located according to the embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. 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.
Some of the words that appear in the text are explained below:
1. the term "and/or" in the embodiments of the present invention describes an association relationship of associated objects, and indicates that three relationships may exist, for example, a and/or B may indicate: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship.
2. In the embodiments of the present application, the term "plurality" means two or more, and other terms are similar.
The technical solutions in the embodiments of the present application will be described below clearly and completely with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, and not all embodiments. 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 application.
It should be noted that the display sequence of the embodiment of the present application only represents the sequence of the embodiment, and does not represent the merits of the technical solutions provided by the embodiments.
Example one
Referring to fig. 1, a schematic diagram of a cargo volume abnormality detection method provided in an embodiment of the present application, as shown in fig. 1, the method includes steps S101 to S1036:
s101, acquiring a second image of a transport tool in a refrigeration house, and adding a label to the second image according to the volume of the transport tool and the offset of the transport tool relative to a door of the refrigeration house to obtain a third image;
s102, training the third image to obtain a transport tool detection model;
s103, collecting a fourth image of goods in the refrigeration house, and adding a label to the fourth image according to the type of the goods box to obtain a fifth image;
s104, training the fifth image to obtain a cargo detection model;
s105, collecting a sixth image of the refrigeration house, and determining a characteristic data set of a freight transportation track according to the sixth image, the transport tool detection model and the cargo detection model;
and S106, detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track.
As a preferred example, in the embodiment of the present invention, before S101, the method may further include:
acquiring size (x) of monitoring video of cold storage scenew,yh) And position information (x) of the cold storage doord,yd,wd,hd)。
Wherein x iswThe length of a monitoring video in a refrigeration house is shown, and the unit is the number of pixels; y ishThe width of the monitoring video in the refrigeration house is the unit of the number of pixels.
Wherein x isdIs the abscissa of the upper left corner of the cold storage door, ydIs the ordinate, w, of the upper left corner of the cold storage doordWidth of the door of the refrigerator, hdThe height of the cold storage door is shown, and the unit is the number of pixels.
As a preferred example, in the embodiment of the present invention, the transportation means includes, but is not limited to, a forklift and a hand truck. For example, in S101 in the embodiment of the present invention, an image (i.e., a second image) of a transportation vehicle such as a forklift and a trolley in a cold storage scene is acquired, and a label is added according to a volume specification of the transportation vehicle and an offset angle of the transportation vehicle with respect to a straight line where a door of the cold storage is located, so as to obtain a third image, where the offset angle of the straight line where the door of the cold storage is located is shown in fig. 5;
as a preferable example, in S102 according to the embodiment of the present invention, the third image data is trained to obtain a vehicle detection model.
Preferably, tagging the second image means adding a vector to the second image, the vector including, but not limited to, the type of vehicle, vehicle identification box coordinates, volume specifications, and the offset angle of the vehicle with respect to the line on which the door of the cold storage dock is located.
In this embodiment of the present invention, the training on the third image may be training based on a YOLO model, or may be other training, and the present invention is not limited in particular.
As a preferable example, in S103 according to the embodiment of the present invention, a fourth image of the goods in the refrigerator is collected, and a label is added to the fourth image according to the type of the goods box, so as to obtain a fifth image. For example, an overall image (i.e., a fourth image) of goods transported in a cold storage scene is acquired, and a fifth image is obtained by adding labels such as cartons and plastic baskets according to box types.
As a preferable example, in S104 in the embodiment of the present invention, the fifth image is trained to obtain a cargo detection model.
Preferably, the labeling of the fourth image in S103 includes labeling the fourth image with a container type label, such as a carton, a plastic box.
It should be noted that, in the embodiment of the present invention, the training on the fifth image may be: the training based on the YOLO model may be other training, and the present invention is not limited specifically.
As a preferable example, in this embodiment S105, determining a characteristic data set of the freight transportation trajectory as shown in fig. 2 includes:
s201, detecting the transport information in the sixth image by using the transport detection model;
s202, if a transport tool exists in the sixth image, detecting the cargo information in the sixth image by using the cargo detection model;
as a preferred example, the step may further include returning to S201 if the transport does not exist in the sixth image.
S203, determining a transport means feature vector set according to the transport means information and the cargo information;
s204, determining a difference matching score of the transport tool according to a transport tool feature vector set obtained from the sixth images of the front frame and the rear frame;
s205, determining a continuous track characteristic vector set of the same transport means according to the difference matching score of the transport means;
s206, determining a characteristic data set of the incoming and outgoing transport track according to the transport means characteristic vector set and the continuous track characteristic vector set.
As a preferred example, the set of vehicle feature vectors in the present embodiment is:
Figure BDA0003414198990000111
wherein,
xmis the horizontal axis coordinate of the transportation tool identification box center;
ymis the longitudinal axis coordinate of the vehicle identification box center;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
dstatusis the transport cargo state of the transport;
xcpthe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis the height of the vehicle identification box;
i is the transport number;
wherein the transport cargo state d of the transport meansstatusDetermined according to the following formula:
Figure BDA0003414198990000112
wherein, tsdtDistance threshold, ts, obtained for historical data trainingagTraining the obtained angle threshold value for historical data;
Figure BDA0003414198990000121
Figure BDA0003414198990000122
wherein A ═ 2xcp+wcp-2xgoods-wgoods，
B＝2ycp+hcp-2ygoods-hgoods,
θs2The deviation angle of the transportation tool relative to the straight line of the cold storage platform door is detected,
ε is a non-negative correction constant and satisfies ε < 1.
As a preferable example, in S205 according to this embodiment of the present invention, determining a set of continuous trajectory feature vectors of the same vehicle according to the difference matching score of the vehicle includes:
determining images of the same transport means according to the difference matching score;
combining the feature vectors of the images of the same transport means to obtain a continuous track feature vector set;
said determining images of the same vehicle from the discrepancy match scores comprises:
vehicle feature vector from previous frame
Figure BDA0003414198990000123
And the vehicle feature vector of the next frame
Figure BDA0003414198990000124
Calculating a disparity match score dij：
Figure BDA0003414198990000125
Wherein, cd1，cd2，cd3The weighting coefficient is obtained by training according to historical data;
when d isij＜tsdfDetermining that the transport means of the front frame and the rear frame are the same;
wherein, tsdfIs a preset threshold value.
As a preferable example, in S206 of the embodiment of the present invention, determining the feature data set of the incoming transportation trajectory according to the vehicle feature vector set and the continuous trajectory feature vector set includes:
calculating the freight state score of each continuous track, and judging that the continuous track is a freight track or an air freight track according to the freight state score;
extracting freight track data in the freight track, and judging whether the track is a freight input transport track or a freight output transport track;
extracting characteristic data of the goods input track to obtain a characteristic data set of the goods input and output track;
wherein the shipment status score is calculated according to the following formula:
Figure BDA0003414198990000131
wherein n isctThe number of elements in the feature set;
if d isct＞tsctJudging the freight track, otherwise, judging the freight track is the air freight track.
Specifically, extracting the freight trajectory data in the freight trajectory, and determining whether the trajectory is a freight incoming transportation trajectory or a freight outgoing transportation trajectory includes:
for the ith feature vector
Figure BDA0003414198990000132
Calculated according to the following formula
Figure BDA0003414198990000133
And
Figure BDA0003414198990000134
equation 1:
Figure BDA0003414198990000135
equation 2:
Figure BDA0003414198990000136
wherein x isdIs the coordinate of the horizontal axis of the endpoint at the upper left corner of the cold storage door frame, ydIs the coordinate of the longitudinal axis of the endpoint at the upper left corner of the cold storage door frame, wdWidth of the door frame of the cold storagedIs the height, x of the door frame of the cold storagewIs the width of the cold storage screen, yhThe picture of the cold storage is high;
according to the formula 1 and theEquation 2 determines the feature vector of the transport at the beginning of the trajectory
Figure BDA0003414198990000137
Is/are as follows
Figure BDA0003414198990000138
And
Figure BDA0003414198990000139
determining a feature vector of a vehicle at the end of a trajectory according to said equations 1 and 2
Figure BDA00034141989900001310
Is/are as follows
Figure BDA00034141989900001311
And
Figure BDA00034141989900001312
if it is
Figure BDA00034141989900001313
And is
Figure BDA00034141989900001314
Judging as an incoming and outgoing freight track;
if it is
Figure BDA00034141989900001315
And is
Figure BDA00034141989900001316
Judging as an incoming and outgoing freight track;
wherein, tsbgAnd tsdgTraining the obtained threshold value for the historical data, and satisfying the following conditions:
Figure BDA00034141989900001317
where min () represents the minimum value.
As a preferable example, in S101 according to the embodiment of the present invention, the detecting whether the cargo volume is abnormal according to the characteristic data set of the cargo transportation track includes:
estimating the volume V of goods transported on each goods-feeding track according to the characteristic data set of the goods-transporting trackiI is the number of the feeding track;
determining a total volume estimate for the cargo according to the following formula:
Vg＝∑Vi；
determining the volume V' of the goods and the quantity N of the goods according to the registration information of the goods, and calculating a registration value of the volume of the goods:
Vr＝V’N；
calculating a volume difference ratio P based on said estimated total volume and said registered volumed：
Figure BDA0003414198990000141
If b isd≤Pd≤buIf the volume of the goods is normal, otherwise, the volume of the goods is abnormal;
wherein, bdAnd buThe volume lower bound threshold and the volume upper bound threshold are obtained by training according to historical data.
Specifically, the estimation of the volume V of the goods transported in each goods-entering track according to the characteristic data set of the goods-transporting trackiThe method comprises the following steps:
in the feature data set, the feature data is (x)cp,ycp,wcp,hcp,xgoods,ygoods,wgoods,hgoods,θ)；
When in use
Figure BDA0003414198990000142
The estimated cargo volume is:
Vi＝hcwclcPwlPh；
when in use
Figure BDA0003414198990000143
The estimated cargo volume is:
Figure BDA0003414198990000144
wherein x iscpThe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis the height of the vehicle identification box;
xgoodsis the horizontal axis coordinate of the endpoint of the upper left corner of the goods square frame;
ygoodsis the coordinates of the vertical axis of the endpoint of the upper left corner of the cargo square;
wgoodsis the width of the cargo box;
hgoodsis the height of the cargo box;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
hcis the vehicle height;
wcis the vehicle cross-sectional width;
lcis the tool length;
Pwlis an estimate of the floor zoom ratio of the good;
Phis an estimate of the cargo's height scaling ratio;
cvis a correction constant obtained by historical data training;
c0is the correction constant.
Specifically, the bottom surface scaling ratio estimate P of the goodwlDetermined by the following equation:
Figure BDA0003414198990000151
wherein, cwwIs a cross section scaling constant obtained by historical data training.
In particular, the estimate of the height scaling ratio of the good PhDetermined by the following equation:
Figure BDA0003414198990000152
wherein, chlIs the length scaling of the relative heights obtained by historical data training, chwWidth scaling of the relative height.
According to the method, firstly, images of a transport tool are collected and trained to obtain a detection model of the transport tool; then, collecting images of the goods, and training to obtain a goods detection model; and then acquiring an image of the refrigeration house, and detecting whether the cargo volume is abnormal or not according to the transport tool detection model and the cargo detection model. The cargo volume abnormity detection method provided by the invention can effectively detect the transported cargo volume in real time, compares the detected cargo volume with the registration information to detect whether the cargo volume is abnormal or not, and compared with the traditional manual detection method, the method has the advantages that the analysis is carried out by using the monitoring video data, the real-time and high-efficiency are realized, the cargo volume abnormity is automatically detected while the cargo handling is not influenced, the detection efficiency is improved, the 24-hour full-coverage detection is ensured, the human resource is saved, the applicability is strong, and the image of the abnormal cargo volume can be reserved to provide technical support for the subsequent inspection decision.
Example two
Based on the same inventive concept, an embodiment of the present invention further provides a cargo volume abnormality detection apparatus, as shown in fig. 3, the apparatus includes:
the data acquisition module 301 is configured to acquire a second image of a transport tool in the refrigerator, a fourth image of goods and a sixth image of the refrigerator;
a cargo volume anomaly detection module 302 configured to tag the second image according to the volume of the vehicle and the offset of the vehicle relative to the cold store door, resulting in a third image; training the third image to obtain the transport tool detection model; adding a label to the fourth image according to the type of the cargo box to obtain a fifth image; training the fifth image to obtain a cargo detection model; determining a characteristic data set of a freight transportation track according to the sixth image, the transport detection model and the cargo detection model; and detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track.
As a preferred example, the cargo volume anomaly detection module 302 is further configured to determine a characteristic data set of the freight transportation trajectory according to the following:
detecting vehicle information in the sixth image using the vehicle detection model;
if the transport means exists in the sixth image, detecting the cargo information in the sixth image by using the cargo detection model;
determining a vehicle characteristic vector set according to the vehicle information and the cargo information;
determining a difference matching score of the transport means according to a transport means feature vector set obtained from the sixth images of the front frame and the rear frame;
determining a set of continuous trajectory feature vectors of the same transport means according to the difference matching score of the transport means;
and determining a characteristic data set of the incoming transportation track according to the transportation tool characteristic vector set and the continuous track characteristic vector set.
The set of vehicle feature vectors is:
Figure BDA0003414198990000171
wherein,
xmin the vehicle identification boxThe horizontal axis coordinates of the heart;
ymis the longitudinal axis coordinate of the vehicle identification box center;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
dstatusis the transport cargo state of the transport;
xcpthe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis the height of the vehicle identification box;
i is the transport number;
wherein the transport cargo state d of the transport meansstatusDetermined according to the following formula:
Figure BDA0003414198990000172
wherein, tsdtDistance threshold, ts, obtained for historical data trainingagTraining the obtained angle threshold value for historical data;
Figure BDA0003414198990000173
Figure BDA0003414198990000181
wherein A ═ 2xcp+wcp-2xgoods-wgoods，
B＝2ycp+hcp-2ygoods-hgoods,
θs2The deviation angle of the transportation tool relative to the straight line of the cold storage platform door is detected,
ε is a non-negative correction constant and satisfies ε < 1.
As a preferred example, the cargo volume anomaly detection module 302 is further configured to determine a set of continuous trajectory feature vectors for the same vehicle by:
determining images of the same transport means according to the difference matching score;
combining the feature vectors of the images of the same transport means to obtain a continuous track feature vector set;
said determining images of the same vehicle from the discrepancy match scores comprises:
vehicle feature vector from previous frame
Figure BDA0003414198990000182
And the vehicle feature vector of the next frame
Figure BDA0003414198990000183
Calculating a disparity match score dij：
Figure BDA0003414198990000184
Wherein, cd1，cd2，cd3The weighting coefficient is obtained by training according to historical data;
when d isij＜tsdfDetermining that the transport means of the front frame and the rear frame are the same;
wherein, tsdfIs a preset threshold value.
As a preferred example, the cargo volume anomaly detection module 302 is further configured to determine a feature data set of the incoming transportation trajectory from the set of transportation tool feature vectors and the set of continuous trajectory feature vectors:
calculating the freight state score of each continuous track, and judging that the continuous track is a freight track or an air freight track according to the freight state score;
extracting freight track data in the freight track, and judging whether the track is a freight input transport track or a freight output transport track;
extracting characteristic data of the goods input track to obtain a characteristic data set of the goods input and output track;
wherein the shipment status score is calculated according to the following formula:
Figure BDA0003414198990000191
wherein n isctThe number of elements in the feature set;
if d isct＞tsctJudging the freight track, otherwise, judging the freight track is the air freight track.
As a preferred example, the cargo volume anomaly detection module 302 is further configured to extract the shipment track data in the shipment track, and determine that the track is an incoming shipment transportation track or an outgoing shipment transportation track:
for the ith feature vector
Figure BDA0003414198990000192
Calculated according to the following formula
Figure BDA0003414198990000193
And
Figure BDA0003414198990000194
equation 1:
Figure BDA0003414198990000195
equation 2:
Figure BDA0003414198990000196
wherein x isdIs the coordinate of the horizontal axis of the endpoint at the upper left corner of the cold storage door frame, ydIs the coordinate of the longitudinal axis of the endpoint at the upper left corner of the cold storage door frame, wdWidth of the door frame of the cold storagedIs the height, x of the door frame of the cold storagewIs a cold storage pictureWidth of (y)hThe picture of the cold storage is high;
determining the characteristic vector of the transportation tool at the beginning of the track according to the formula 1 and the formula 2
Figure BDA0003414198990000197
Is/are as follows
Figure BDA0003414198990000198
And
Figure BDA0003414198990000199
determining a feature vector of a vehicle at the end of a trajectory according to said equations 1 and 2
Figure BDA00034141989900001910
Is/are as follows
Figure BDA00034141989900001911
And
Figure BDA00034141989900001912
if it is
Figure BDA00034141989900001913
And is
Figure BDA00034141989900001914
Judging as an incoming and outgoing freight track;
if it is
Figure BDA00034141989900001915
And is
Figure BDA00034141989900001916
Judging as an incoming and outgoing freight track;
wherein, tsbgAnd tsdgTraining the obtained threshold value for the historical data, and satisfying the following conditions:
Figure BDA00034141989900001917
where min () represents the minimum value.
As a preferred example, the cargo volume anomaly detection module 302 is further configured to detect whether the cargo volume is anomalous:
estimating the volume V of goods transported on each goods-feeding track according to the characteristic data set of the goods-transporting trackiI is the number of the feeding track;
determining a total volume estimate for the cargo according to the following formula:
Vg＝∑Vi；
determining the volume V' of the goods and the quantity N of the goods according to the registration information of the goods, and calculating a registration value of the volume of the goods:
Vr＝V’N；
calculating a volume difference ratio P based on said estimated total volume and said registered volumed：
Figure BDA0003414198990000201
If b isd≤Pd≤buIf the volume of the goods is normal, otherwise, the volume of the goods is abnormal;
wherein, bdAnd buThe volume lower bound threshold and the volume upper bound threshold are obtained by training according to historical data.
As a preferred example, the cargo volume anomaly detection module 302 is further configured to estimate the volume V of the transported cargo per incoming tracki：
In the feature data set, the feature data is (x)cp,ycp,wcp,hcp,xgoods,ygoods,wgoods,hgoods,θ)；
When in use
Figure BDA0003414198990000202
The estimated cargo volume is：
Vi＝hcwclcPwlPh；
When in use
Figure BDA0003414198990000203
The estimated cargo volume is:
Figure BDA0003414198990000204
wherein x iscpThe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis the height of the vehicle identification box;
xgoodsis the horizontal axis coordinate of the endpoint of the upper left corner of the goods square frame;
ygoodsis the coordinates of the vertical axis of the endpoint of the upper left corner of the cargo square;
wgoodsis the width of the cargo box;
hgoodsis the height of the cargo box;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
hcis the vehicle height;
wcis the vehicle cross-sectional width;
lcis the tool length;
Pwlis an estimate of the floor zoom ratio of the good;
Phis an estimate of the cargo's height scaling ratio;
cvis a correction constant obtained by historical data training;
c0is the correction constant.
In particular, the bottom zoom ratio estimate P of the goodwlIs determined by the following formula：
Figure BDA0003414198990000211
Wherein, cwwIs a cross section scaling constant obtained by historical data training.
In particular, the estimate P of the cargo's height scaling ratiohDetermined by the following equation:
Figure BDA0003414198990000212
wherein, chlIs the length scaling of the relative heights obtained by historical data training, chwWidth scaling of the relative height.
It should be noted that the cargo volume abnormality detection module 302 provided in this embodiment can implement all the functions included in steps S102 to S106 in the first embodiment, solve the same technical problem, achieve the same technical effect, and is not described herein again;
it should be noted that the apparatus provided in the second embodiment and the method provided in the first embodiment belong to the same inventive concept, the same technical problems are solved, the same technical effects are achieved, and the apparatus provided in the second embodiment can implement all the methods of the first embodiment, and the same parts are not described again.
EXAMPLE III
Based on the same inventive concept, an embodiment of the present invention further provides a cargo volume abnormality detection apparatus, as shown in fig. 4, the apparatus includes:
including memory 402, processor 401, and user interface 403;
the memory 402 for storing a computer program;
the user interface 403 is used for realizing interaction with a user;
the processor 401 is configured to read the computer program in the memory 402, and when the processor 401 executes the computer program, the processor implements:
acquiring a second image of a transport tool in a refrigeration house, and adding a label to the second image according to the volume of the transport tool and the offset of the transport tool relative to a door of the refrigeration house to obtain a third image;
training the third image to obtain the transport tool detection model;
acquiring a fourth image of goods in a refrigeration house, and adding a label to the fourth image according to the type of a goods box to obtain a fifth image;
training the fifth image to obtain a cargo detection model;
collecting a sixth image of the refrigeration house, and determining a characteristic data set of a freight transportation track according to the sixth image, the transport tool detection model and the cargo detection model;
and detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track.
Where in fig. 4 the bus architecture may include any number of interconnected buses and bridges, in particular one or more processors, represented by processor 401, and various circuits of memory, represented by memory 402, linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The processor 401 is responsible for managing a bus architecture and general processing, and the memory 402 may store data used by the processor 401 in performing operations.
The processor 401 may be a CPU, ASIC, FPGA or CPLD, and the processor 401 may also employ a multi-core architecture.
When the processor 401 executes the computer program stored in the memory 402, it implements the cargo volume abnormality detection method according to any one of the first embodiment.
It should be noted that the apparatus provided in the third embodiment and the method provided in the first embodiment belong to the same inventive concept, and solve the same technical problem to achieve the same technical effect.
The present application also proposes a processor-readable storage medium. The processor-readable storage medium stores a computer program, and the processor executes the computer program to implement any cargo volume abnormality detection method in the first embodiment.
It should be noted that the division of the cells in the embodiment of the present application is schematic, and is only a logic function division, and there may be another division manner in actual implementation. In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware or a form of a software functional unit.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application 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 application has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. 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.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well. .

Claims (12)

1. A cargo volume abnormality detection method is characterized by comprising:
acquiring a second image of a transport tool in a refrigeration house, and adding a label to the second image according to the volume of the transport tool and the offset of the transport tool relative to a door of the refrigeration house to obtain a third image;
training the third image to obtain the transport tool detection model;
acquiring a fourth image of goods in a refrigeration house, and adding a label to the fourth image according to the type of a goods box to obtain a fifth image;
training the fifth image to obtain a cargo detection model;
collecting a sixth image of the refrigeration house, and determining a characteristic data set of a freight transportation track according to the sixth image, the transport tool detection model and the cargo detection model;
and detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track.
2. The method of claim 1, wherein determining a characteristic data set for a shipment trajectory from the sixth image, the vehicle detection model, and the cargo detection model comprises:
detecting vehicle information in the sixth image using the vehicle detection model;
if the transport means exists in the sixth image, detecting the cargo information in the sixth image by using the cargo detection model;
determining a vehicle characteristic vector set according to the vehicle information and the cargo information;
determining a difference matching score of the transport means according to a transport means feature vector set obtained from the sixth images of the front and rear frames;
determining a set of continuous trajectory feature vectors of the same transport means according to the difference matching score of the transport means;
and determining a characteristic data set of the incoming and outgoing transport track according to the transport means characteristic vector set and the continuous track characteristic vector set.
3. The method of claim 2, wherein the set of vehicle feature vectors is:
Figure FDA0003414198980000021
wherein,
xmis the horizontal axis coordinate of the transportation tool identification box center;
ymis the longitudinal axis coordinate of the vehicle identification box center;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
dstatusis the transport cargo state of the transport;
xcpthe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis a means of transportIdentifying a high of the square;
i is the transport number;
wherein the transport cargo state d of the transport meansstatusDetermined according to the following formula:
Figure FDA0003414198980000022
wherein, tsdtDistance threshold, ts, obtained for historical data trainingagTraining the obtained angle threshold value for historical data;
Figure FDA0003414198980000023
Figure FDA0003414198980000024
wherein A ═ 2xcp+wcp-2xgoods-wgoods，
B＝2ycp+hcp-2ygoods-hgoods,
θs2The deviation angle of the transportation tool relative to the straight line of the cold storage platform door is detected,
ε is a non-negative correction constant and satisfies ε < 1.
4. The method of claim 2, wherein determining a set of consecutive trajectory feature vectors for the same vehicle based on the vehicle's variance match scores comprises:
determining images of the same transport means according to the difference matching score;
combining the feature vectors of the images of the same transport means to obtain a continuous track feature vector set;
said determining images of the same vehicle from the discrepancy match scores comprises:
vehicle feature vector from previous frame
Figure FDA0003414198980000031
And the vehicle feature vector of the next frame
Figure FDA0003414198980000032
Calculating a disparity match score dij：
Figure FDA0003414198980000033
Wherein, cd1，cd2，cd3The weighting coefficient is obtained by training according to historical data;
when d isij＜tsdfDetermining that the transport means of the front frame and the rear frame are the same;
wherein, tsdfIs a preset threshold value.
5. The method of claim 2, wherein determining a feature data set of an in-freight transportation trajectory from the set of vehicle feature vectors and the set of continuous trajectory feature vectors comprises:
calculating the freight state score of each continuous track, and judging whether the continuous track is a freight track or an air freight track according to the freight state score;
extracting freight track data in the freight track, and judging whether the track is a freight input transport track or a freight output transport track;
extracting characteristic data of the goods input track to obtain a characteristic data set of the goods input and output track;
wherein the shipment status score is calculated according to the following formula:
Figure FDA0003414198980000034
wherein，nctThe number of elements in the feature set;
if d isct＞tsctJudging the freight track, otherwise, judging the freight track is the air freight track.
6. The method of claim 5, wherein the extracting of the shipment trajectory data in the shipment trajectory and the determining of the trajectory as a shipment transportation trajectory or a shipment transportation trajectory comprises:
for the ith feature vector
Figure FDA0003414198980000041
Calculated according to the following formula
Figure FDA0003414198980000042
And
Figure FDA0003414198980000043
equation 1:
Figure FDA0003414198980000044
equation 2:
Figure FDA0003414198980000045
wherein x isdIs the coordinate of the horizontal axis of the endpoint at the upper left corner of the cold storage door frame, ydIs the coordinate of the longitudinal axis of the endpoint at the upper left corner of the cold storage door frame, wdWidth of the door frame of the cold storagedIs the height, x of the door frame of the cold storagewIs the width of the cold storage framehThe picture of the cold storage is high;
determining the characteristic vector of the transportation tool at the beginning of the track according to the formula 1 and the formula 2
Figure FDA0003414198980000046
Is/are as follows
Figure FDA0003414198980000047
And
Figure FDA0003414198980000048
determining a feature vector of a vehicle at the end of a trajectory according to said equations 1 and 2
Figure FDA0003414198980000049
Is/are as follows
Figure FDA00034141989800000410
And
Figure FDA00034141989800000411
if it is
Figure FDA00034141989800000412
And is
Figure FDA00034141989800000413
Judging as an incoming and outgoing freight track;
if it is
Figure FDA00034141989800000414
And is
Figure FDA00034141989800000415
Judging as an incoming and outgoing freight track;
wherein, tsbgAnd tsdgTraining the obtained threshold value for the historical data, and satisfying the following conditions:
Figure FDA00034141989800000416
where min () represents the minimum value.
7. The method of claim 1, wherein the detecting whether the cargo volume is abnormal from the set of characteristic data of the cargo transportation trajectory comprises:
estimating the volume V of goods transported on each goods-feeding track according to the characteristic data set of the goods-transporting trackiI is the number of the feeding track;
determining a total volume estimate for the cargo according to the following formula:
Vg＝∑Vi；
determining the volume V' of the goods and the quantity N of the goods according to the registration information of the goods, and calculating a registration value of the volume of the goods:
Vr＝V’N；
calculating a volume difference ratio P based on said estimated total volume and said registered volumed：
Figure FDA0003414198980000051
If b isd≤Pd≤buIf the volume of the goods is normal, otherwise, the volume of the goods is abnormal;
wherein, bdAnd buThe volume lower bound threshold and the volume upper bound threshold are obtained by training according to historical data.
8. The method of claim 7, wherein estimating the volume V of the shipment per incoming track based on the set of characteristic data for the shipment trackiThe method comprises the following steps:
in the feature data set, the feature data is (x)cp,ycp,wcp,hcp,xgoods,ygoods,wgoods,hgoods,θ)；
When in use
Figure FDA0003414198980000052
The estimated cargo volume is:
Vi＝hcwclcPwlPh；
when in use
Figure FDA0003414198980000053
The estimated cargo volume is:
Figure FDA0003414198980000054
wherein x iscpThe coordinates of the horizontal axis of the endpoint of the upper left corner of the transportation tool identification frame;
ycpis the top left corner endpoint longitudinal axis coordinate of the vehicle identification box;
wcpis the width of the vehicle identification box;
hcpis the height of the vehicle identification box;
xgoodsis the horizontal axis coordinate of the endpoint of the upper left corner of the goods square frame;
ygoodsis the coordinates of the vertical axis of the endpoint of the upper left corner of the cargo square;
wgoodsis the width of the cargo box;
hgoodsis the height of the cargo box;
theta is the offset angle of the transportation tool relative to the straight line where the dock door of the refrigeration house is located;
hcis the vehicle height;
wcis the vehicle cross-sectional width;
lcis the tool length;
Pwlis an estimate of the floor zoom ratio of the good;
Phis an estimate of the cargo's height scaling ratio;
cvis a correction constant obtained by historical data training;
c0is the correction constant.
9. The method of claim 8, wherein the floor scaling estimate P of the good is an estimate of the floor scaling ratio of the goodwlDetermined by the following equation:
Figure FDA0003414198980000061
wherein, cwwIs a cross section scaling constant obtained by historical data training.
10. The method of claim 8, wherein the cargo height scaling ratio estimate PhDetermined by the following equation:
Figure FDA0003414198980000062
wherein, chlIs the length scaling of the relative heights obtained by historical data training, chwWidth scaling of the relative height.
11. An abnormal cargo volume detection device, comprising:
the data acquisition module is configured for acquiring a second image of the transport tool in the refrigeration house, a fourth image of the goods and a sixth image of the refrigeration house;
a cargo volume anomaly detection module configured to tag the second image according to the volume of the transport and the offset of the transport relative to the cold store door to obtain a third image; training the third image to obtain the transport tool detection model; adding a label to the fourth image according to the type of the cargo box to obtain a fifth image; training the fifth image to obtain a cargo detection model; determining a characteristic data set of a freight transportation track according to the sixth image, the transport detection model and the cargo detection model; and detecting whether the cargo volume is abnormal or not according to the characteristic data set of the cargo transportation track.
12. The cargo volume abnormity detection device is characterized by comprising a memory, a processor and a user interface;
the memory for storing a computer program;
the user interface is used for realizing interaction with a user;
the processor, configured to read the computer program in the memory, and when the processor executes the computer program, implement the cargo volume abnormality detection method according to one of claims 1 to 10.
CN202111540672.8A 2021-12-16 2021-12-16 Cargo volume anomaly detection method and device and storage medium Active CN114202700B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111540672.8A CN114202700B (en) 2021-12-16 2021-12-16 Cargo volume anomaly detection method and device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111540672.8A CN114202700B (en) 2021-12-16 2021-12-16 Cargo volume anomaly detection method and device and storage medium

Publications (2)

Publication Number Publication Date
CN114202700A true CN114202700A (en) 2022-03-18
CN114202700B CN114202700B (en) 2022-07-22

Family

ID=80654358

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111540672.8A Active CN114202700B (en) 2021-12-16 2021-12-16 Cargo volume anomaly detection method and device and storage medium

Country Status (1)

Country Link
CN (1) CN114202700B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115600823A (en) * 2022-12-08 2023-01-13 东莞先知大数据有限公司（Cn） Method for detecting abnormality in transported goods, electronic device and storage medium

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100117832A1 (en) * 2008-11-07 2010-05-13 Electronics And Telecommunication Research Institute System and method for managing information of loading goods
CN109003390A (en) * 2018-06-29 2018-12-14 深圳和而泰数据资源与云技术有限公司 A kind of commodity recognition method, self-service machine and computer readable storage medium
CN110363463A (en) * 2019-06-04 2019-10-22 天津五八到家科技有限公司 Ship data processing method, equipment, system and storage medium
CN110905328A (en) * 2018-09-14 2020-03-24 浙江宇视科技有限公司 Quick access door control method and quick access door control device
US20200104785A1 (en) * 2018-09-28 2020-04-02 I.D. Systems, Inc. Cargo sensors, cargo-sensing units, cargo-sensing systems, and methods of using the same
CN111340415A (en) * 2020-02-14 2020-06-26 北京航空航天大学 Vehicle spontaneous formation mode mining method based on freight track data
CN111645730A (en) * 2020-06-16 2020-09-11 武汉钢铁有限公司 Locomotive operation efficiency analysis system
CN112639852A (en) * 2018-08-27 2021-04-09 戴姆勒股份公司 Method and system for vehicle cargo management
US11004093B1 (en) * 2009-06-29 2021-05-11 Videomining Corporation Method and system for detecting shopping groups based on trajectory dynamics
CN113762862A (en) * 2020-12-28 2021-12-07 京东城市(北京)数字科技有限公司 Cargo abnormity monitoring method and device, electronic equipment and storage medium

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100117832A1 (en) * 2008-11-07 2010-05-13 Electronics And Telecommunication Research Institute System and method for managing information of loading goods
US11004093B1 (en) * 2009-06-29 2021-05-11 Videomining Corporation Method and system for detecting shopping groups based on trajectory dynamics
CN109003390A (en) * 2018-06-29 2018-12-14 深圳和而泰数据资源与云技术有限公司 A kind of commodity recognition method, self-service machine and computer readable storage medium
CN112639852A (en) * 2018-08-27 2021-04-09 戴姆勒股份公司 Method and system for vehicle cargo management
CN110905328A (en) * 2018-09-14 2020-03-24 浙江宇视科技有限公司 Quick access door control method and quick access door control device
US20200104785A1 (en) * 2018-09-28 2020-04-02 I.D. Systems, Inc. Cargo sensors, cargo-sensing units, cargo-sensing systems, and methods of using the same
CN110363463A (en) * 2019-06-04 2019-10-22 天津五八到家科技有限公司 Ship data processing method, equipment, system and storage medium
CN111340415A (en) * 2020-02-14 2020-06-26 北京航空航天大学 Vehicle spontaneous formation mode mining method based on freight track data
CN111645730A (en) * 2020-06-16 2020-09-11 武汉钢铁有限公司 Locomotive operation efficiency analysis system
CN113762862A (en) * 2020-12-28 2021-12-07 京东城市(北京)数字科技有限公司 Cargo abnormity monitoring method and device, electronic equipment and storage medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
杨树玉 等: ""一种基于立式突出路标的巡检机器人视觉自定位方法"", 《南京师范大学学报(工程技术版)》 *
潘超 等: ""昆虫视觉启发的光流复合导航方法"", 《自动化学报》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115600823A (en) * 2022-12-08 2023-01-13 东莞先知大数据有限公司（Cn） Method for detecting abnormality in transported goods, electronic device and storage medium

Also Published As

Publication number Publication date
CN114202700B (en) 2022-07-22

Similar Documents

Publication Publication Date Title
EP3620981B1 (en) Object detection method, device, apparatus and computer-readable storage medium
CN114529864B (en) Method and device for detecting shoreside smuggling behavior and storage medium
CN111310645B (en) Method, device, equipment and storage medium for warning overflow bin of goods accumulation
CN110852162B (en) Human body integrity data labeling method and device and terminal equipment
WO2021151277A1 (en) Method and apparatus for determining severity of damage on target object, electronic device, and storage medium
CN114202700B (en) Cargo volume anomaly detection method and device and storage medium
CN107403128B (en) Article identification method and device
CN112037177B (en) Carriage loading rate assessment method and device and storage medium
Saran et al. Robust visual analysis for planogram compliance problem
CN113392681A (en) Human body falling detection method and device and terminal equipment
CN114758464A (en) Storage battery anti-theft method, device and storage medium based on charging pile monitoring video
US20230326061A1 (en) A Cargo Tracking System for Tracking a Cargo, as Well as a Corresponding Method
CN111444796B (en) Goods placement judging method and device of vending robot
CN117132872A (en) Intelligent collision recognition system for material transport vehicle on production line
CN112686162A (en) Method, device, equipment and storage medium for detecting clean state of warehouse environment
CN115471787A (en) Construction site object stacking detection method and device and storage medium
CN114445725A (en) Method and device for detecting fresh goods
CN111401104B (en) Classification model training method, classification method, device, equipment and storage medium
CN115171061A (en) Vehicle detection method, device and equipment
CN115410105A (en) Container mark identification method, device, computer equipment and storage medium
CN114627430B (en) Taking takeout detection method and device and storage medium
CN113313968A (en) Parking space detection method and storage medium
CN117405124B (en) Path planning method and system based on big data
WO2024017413A2 (en) Method and apparatus for port entry/exit detection of vehicle, device, and storage medium
CN115482538B (en) Material label extraction method and system based on Mask R-CNN

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
CP03 Change of name, title or address
CP03 Change of name, title or address

Address after: Building 7, No. 124 Dongbao Road, Dongcheng Street, Dongguan City, Guangdong Province, 523128

Patentee after: Guangdong Prophet Big Data Co.,Ltd.

Country or region after: China

Address before: 523128 Room 401, building 6, No.5 Weifeng Road, Dongcheng Street, Dongguan City, Guangdong Province

Patentee before: Dongguan prophet big data Co.,Ltd.

Country or region before: China