CN112446896A - Conveying material falling monitoring method, device and system based on image recognition - Google Patents

Conveying material falling monitoring method, device and system based on image recognition Download PDF

Info

Publication number
CN112446896A
CN112446896A CN202110132859.8A CN202110132859A CN112446896A CN 112446896 A CN112446896 A CN 112446896A CN 202110132859 A CN202110132859 A CN 202110132859A CN 112446896 A CN112446896 A CN 112446896A
Authority
CN
China
Prior art keywords
image
falling
boundary
distance
probability
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
CN202110132859.8A
Other languages
Chinese (zh)
Other versions
CN112446896B (en
Inventor
李政德
刘霞
武杰
戴冬冬
霍英杰
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.)
Austong Intelligent Robot Technology Co Ltd
Original Assignee
Austong Intelligent Robot 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 Austong Intelligent Robot Technology Co Ltd filed Critical Austong Intelligent Robot Technology Co Ltd
Priority to CN202110132859.8A priority Critical patent/CN112446896B/en
Publication of CN112446896A publication Critical patent/CN112446896A/en
Application granted granted Critical
Publication of CN112446896B publication Critical patent/CN112446896B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G43/00—Control devices, e.g. for safety, warning or fault-correcting
    • B65G43/08—Control devices operated by article or material being fed, conveyed or discharged
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G06T7/248—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments involving reference images or patches
    • 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
    • G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/18—Status alarms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/61—Control of cameras or camera modules based on recognised objects
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G2203/00—Indexing code relating to control or detection of the articles or the load carriers during conveying
    • B65G2203/02—Control or detection
    • B65G2203/0208—Control or detection relating to the transported articles
    • B65G2203/0225—Orientation of the article
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G2203/00—Indexing code relating to control or detection of the articles or the load carriers during conveying
    • B65G2203/02—Control or detection
    • B65G2203/0208—Control or detection relating to the transported articles
    • B65G2203/025—Speed of the article
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G2203/00—Indexing code relating to control or detection of the articles or the load carriers during conveying
    • B65G2203/04—Detection means
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Control Of Conveyors (AREA)

Abstract

The embodiment of the invention relates to the technical field of truck-mounted material conveying, and discloses a method, a device and a system for monitoring falling of conveyed materials based on image recognition, wherein a first image is detected, a first side edge of a conveying belt and a first boundary of the materials are recognized, and a first distance between the first side edge and the first boundary is calculated; judging whether the first distance is smaller than a first distance threshold value, if so, acquiring a second image set of the material, and calculating the movement speed and the movement direction of the material; calculating the falling probability of the materials based on the moving speed, the moving direction and the first distance of the materials; and if the falling probability is greater than the first probability threshold, sending a material falling warning. The invention can automatically identify the falling state of the material on the conveying belt, improves the identification speed of the falling state of the material based on the image and reduces the calculation amount of the falling monitoring of the material; on the basis of accurately identifying the falling state of the material and alarming the falling state, different processing can be performed on different material states, and the risk of falling of the material is reduced.

Description

Conveying material falling monitoring method, device and system based on image recognition
Technical Field
The embodiment of the invention relates to the technical field of truck loading material conveying, in particular to a conveying material falling monitoring method, device and system based on image recognition.
Background
The inventor finds that at least the following problems exist in the prior art: along with the improvement of loading and transporting efficiency, the supply of materials is basically realized by the conveyer belt. However, the conveying belt conveys materials to the loading position and needs to pass through a longer conveying path, the materials on the conveying belt easily fall off in the conveying path, the materials are easily lost if the conveying belt is unattended, and the loading automation degree is reduced if people are sent for a long time on duty, so that the loading cost is improved, and the dependence on manpower is improved. Because the falling of the material has the burstiness and the instantaneity, the falling event of the material is usually monitored by adopting an infrared shielding principle in the prior art, but the infrared monitoring cannot distinguish whether the shielding object is the falling material or other objects, so that the false alarm phenomenon is easy to occur; the invention patent with application number CN201810712357.0 provides a method for determining material falling based on image recognition, however, in the method, it is necessary to recognize the characteristic information of the conveyor belt based on the image to determine whether the material falls, the implementation of the method depends on the information such as color and characteristic pattern of the conveyor belt, i.e. on the specific conveyor belt, which is not beneficial to popularization and use, and the cost and workload for modifying the existing conveyor belt are high. The existing conveying belt conveys materials to the designated position, and the danger that the materials fall off exists in each conveying position, so that the automatic, quick and accurate material falling monitoring scheme is urgently needed.
Disclosure of Invention
The embodiment of the invention aims to provide a method, a device and a system for monitoring falling of conveyed materials based on image recognition, and solves the technical problems that the material falling detection accuracy is low, the equipment modification cost is high, and the material falling cannot be automatically, quickly and accurately monitored in the prior art.
In order to solve the technical problem, an embodiment of the present invention provides a conveying material drop monitoring method based on image recognition, including the following steps:
step S1: collecting a first image of a material;
step S2, detecting the first image, identifying a first side edge of the conveyor belt and a first boundary of the material, and calculating a first distance between the first side edge and the first boundary;
step S3, judging whether the first distance is smaller than a first distance threshold value, if not, returning to the step S1; if yes, go to step S4;
step S4, acquiring a second image set of the material, and calculating the movement speed and the movement direction of the material based on the first image and the second image set;
step S5, calculating the falling probability of the material based on the movement speed, the movement direction and the first distance of the material;
and step S6, if the falling probability is larger than a first probability threshold, sending out a material falling warning.
Preferably, the step S2 specifically includes:
if the first boundary of the material and the first side edge of the conveying belt are the same, the first distance is the distance between the first boundary of the material and the first side edge of the conveying belt;
if a first included angle exists between the first boundary of the material and the first side edge of the conveying belt, calculating the distance between the projection point of the gravity center of the material on the first boundary and the first side edge of the conveying belt to obtain a first distance.
Preferably, the step S4 specifically includes:
the second image set is a historical image set of the material,
step S41, determining an effective image area where the first boundary in the first image is located;
step S42 of determining a history image effective region of each history image in the history image set based on the effective image region;
step S43, determining the material boundary in each history image based on the effective region of the history image;
in step S44, the moving speed and the moving direction of the material are calculated based on the position of the boundary of the material in each history image and the first boundary in the first image.
Preferably, the historical image set includes historical images
Figure 557965DEST_PATH_IMAGE001
，
Figure 475105DEST_PATH_IMAGE002
The number of the history images in the history image set is as follows
Figure 192525DEST_PATH_IMAGE003
Calculating the moving speed of two adjacent images according to the time sequence
Figure 525418DEST_PATH_IMAGE004
Wherein, in the step (A),
Figure 790657DEST_PATH_IMAGE005
is a first image
Figure 613119DEST_PATH_IMAGE003
And history image
Figure 817835DEST_PATH_IMAGE006
To get rid of
Figure 16736DEST_PATH_IMAGE002
The speed of the movement of the boundary is,
Figure 623297DEST_PATH_IMAGE007
as a history image
Figure 554344DEST_PATH_IMAGE006
And Pn-1To get rid of
Figure 744892DEST_PATH_IMAGE008
The speed of the boundary motion, …,
Figure 481904DEST_PATH_IMAGE009
as a history image
Figure 942972DEST_PATH_IMAGE010
And
Figure 310500DEST_PATH_IMAGE011
1 st boundary movement speed;
the movement speed of the materials is as follows:
Figure 552125DEST_PATH_IMAGE012
wherein, in the step (A),
Figure 764932DEST_PATH_IMAGE013
is shown as
Figure 847551DEST_PATH_IMAGE014
The speed of the movement of the boundary is,
Figure 120400DEST_PATH_IMAGE015
is as follows
Figure 849322DEST_PATH_IMAGE014
The influence factor corresponding to the speed of the boundary motion,
Figure 865819DEST_PATH_IMAGE016
，
Figure 301480DEST_PATH_IMAGE017
。
preferably, the step S6 specifically includes:
if the falling probability is larger than a first probability threshold, identifying material information based on a first image, and sending a material falling warning, wherein the material falling warning comprises material information.
Preferably, the step S6 specifically includes:
if the falling probability is smaller than or equal to a first probability threshold and is larger than a second probability threshold, calculating the estimated falling time of the material according to the movement speed, the movement direction and the first distance of the material, and sending a correction signal according to the estimated falling time to correct the movement direction and the movement speed of the material.
Preferably, the step S6 specifically includes:
after the movement direction and the movement speed of the material are corrected, the sampling frequency of the camera is increased, and the step S1 is returned;
and if the falling probability is less than or equal to the second probability threshold, reducing the sampling frequency of the camera, and returning to the step S1.
The embodiment of the invention also provides a conveying material falling monitoring device based on image recognition, which comprises:
the image acquisition module is used for acquiring a first image of the material;
the distance calculation module is used for detecting the first image, identifying a first side edge of the conveying belt and a first boundary of the material, and calculating a first distance between the first side edge and the first boundary;
the judging module is used for judging whether the first distance is smaller than a first distance threshold value;
the movement calculation module is used for acquiring a second image set of the material when the first distance is smaller than a first distance threshold value, and calculating the movement speed and the movement direction of the material based on the first image and the second image set;
the falling probability calculation module is used for calculating the falling probability of the materials based on the movement speed, the movement direction and the first distance of the materials;
and the state processing module is used for sending out a material falling warning if the falling probability is greater than a first probability threshold.
The embodiment of the invention also provides a conveying material falling monitoring system based on image recognition, which is characterized by comprising one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing any of the methods described above.
Compared with the prior art, the embodiment of the invention directly determines the relative position relation between the materials on the conveying belt and the conveying belt based on the image, monitors the falling state of the materials, does not need to modify the conveying belt or the materials, reduces the modification cost of equipment and improves the applicability of the material falling monitoring method; the material falling probability is calculated based on the relative position relation between the material boundary and the side edge of the conveying belt and the movement speed and the movement direction of the material, the interval range to which the material falling probability belongs is distinguished, only the falling state alarm is carried out on the material with the falling probability in the first interval, the accuracy of the material falling state alarm is improved, the material with the falling probability in other intervals is not in the falling state, the material monitoring method can carry out self-adaptive processing, the falling risk of the material is reduced, only the material with the falling risk needs to be calculated according to the speed, the direction and the probability, only the material with the falling probability in the first interval needs to identify material information, the operation amount of the material monitoring method is reduced, and the operation speed of the monitoring method is improved.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
One or more embodiments are illustrated by way of example in the accompanying drawings, which correspond to the figures in which like reference numerals refer to similar elements and which are not to scale unless otherwise specified.
Fig. 1 is a schematic diagram of a conveying material drop monitoring method based on image recognition according to an embodiment of the present invention;
FIG. 2 is a schematic view of a conveyor belt provided by an embodiment of the present invention;
fig. 3 is a top view of a conveyor belt provided by an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, it will be appreciated by those of ordinary skill in the art that numerous technical details are set forth in order to provide a better understanding of the present application in various embodiments of the present invention. However, the technical solution claimed in the present application can be implemented without these technical details and various changes and modifications based on the following embodiments. The following embodiments are divided for convenience of description, and should not constitute any limitation to the specific implementation manner of the present invention, and the embodiments may be mutually incorporated and referred to without contradiction.
A first embodiment of the present invention relates to a method for monitoring dropping of a conveyed material based on image recognition, and as shown in fig. 1, the method for monitoring dropping of a conveyed material based on image recognition specifically includes the following steps:
step S1: a first image of the material is acquired.
The conveyer belt has left and right sides, and left side
Figure 73127DEST_PATH_IMAGE018
And a right side edge
Figure 459983DEST_PATH_IMAGE019
The conveying belt is provided with one or more materials, the materials are conveyed to the loading robot from the material disassembling position under the conveying of the conveying belt, the camera is arranged above the conveying belt, a first image of the materials on the conveying belt is collected, and the first image at least comprises the left side edge and/or the right side edge of the conveying beltThe side edge at least comprises at least one complete material.
Step S2, detecting the first image, identifying a first side of the conveyor belt and a first boundary of the material, and calculating a first distance between the first side and the first boundary.
The method comprises the steps of detecting a first side edge of a conveying belt and a first boundary of materials in a first image, wherein the first side edge is the left side edge or the right side edge of the conveying belt, the first boundary and the material boundary which is closest to the first side edge are the first boundary, for example, the first side edge is the left side edge of the conveying belt, and the first boundary is the material boundary with the material parallel to the left side edge of the conveying belt or the included angle between the material and the left side edge of the conveying belt smaller than a preset. And calculating a first distance between the first side edge and the first boundary based on the identified first side edge and the first boundary of the material, and obtaining the distance between the material and the boundary of the conveying belt.
Step S3, judging whether the first distance is smaller than a first distance threshold value, if not, returning to the step S1; if yes, go to step S4.
According to the method, after the material image is obtained, only the first side edge and the first boundary information need to be detected, compared with the prior art that the material falling state is directly identified based on the image, the method does not need to identify complex state information through the image in the initial state, only the distance relation between the boundary of the material without falling risk and the side edge of the conveying belt needs to be monitored, the steps of material identification and material state identification are not needed, the image identification operand in the material monitoring process is simplified, and the speed of the monitoring method is improved.
Step S4, acquiring a second image set of the material, and calculating the movement speed and the movement direction of the material based on the first image and the second image set;
the second image set is a historical image set of the material, and the moving speed and the moving direction of the material are calculated based on the historical image set and the first image. In the prior art, a general image recognition method is adopted to detect the falling state of the material, in the whole conveying path, the system needs to acquire material images at a higher sampling rate and recognize postures, the calculation amount is higher, and the system overhead is higher.
Step S5, calculating the falling probability of the material based on the movement speed, the movement direction and the first distance of the material;
and step S6, if the falling probability is larger than a first probability threshold, sending out a material falling warning.
In summary, the embodiment provides a conveying material drop monitoring method based on image recognition, the method directly utilizes the image to recognize the position relationship between the material and the side edge of the conveying belt, and does not need to coat a specific pattern or mark on the conveying belt, so that the applicability of the material drop monitoring method is improved, and the cost of improving the conveying belt is reduced. In the image identification process, the falling state of all materials is not detected, the judgment condition related to the first interval is set as the first filtering judgment condition before the falling state is detected and processed, so that only the materials with the falling risk need to be subjected to complex operation based on the image, the materials which do not meet the first filtering condition are regarded as safe materials, and complex image operation is not needed.
The second embodiment of the invention relates to a conveying material falling monitoring method based on image recognition. The second embodiment is substantially the same as the first embodiment, and the same contents as those in the first embodiment are not repeated in this embodiment, and the method for monitoring falling of the conveyed material based on image recognition in the second embodiment includes the following steps:
step S1: a first image of the material is acquired.
As shown in FIGS. 2 and 3, the conveyor belt has left and right sides, and a left side
Figure 342489DEST_PATH_IMAGE018
And a right side edge
Figure 632656DEST_PATH_IMAGE019
With one or more materials, e.g. materials, placed on a conveyor belt
Figure 512887DEST_PATH_IMAGE020
And materials
Figure 950822DEST_PATH_IMAGE021
Of materials
Figure 309122DEST_PATH_IMAGE020
The two boundaries and the corresponding side edges of the conveyer belt form included angles, and the materials are
Figure 220839DEST_PATH_IMAGE021
The two boundaries are parallel to the corresponding side edges of the conveyer belt.
Step S2, detecting the first image, identifying a first side of the conveyor belt and a first boundary of the material, and calculating a first distance between the first side and the first boundary.
The method comprises the steps of firstly calculating the relative position relation between the conveying belt and the materials based on the images, and calculating the distance between the side edge of the conveying belt and the material boundary so as to judge whether the materials have the falling risk. Specifically, the first side is the right side, detects first image, discerns the right side of conveyer belt, simultaneously, detects the first border of material, and first border is the right border of material equally. If the right material boundary is parallel to the right side of the conveyer belt, the first distance is the distance between the right material boundary and the right side of the conveyer belt
Figure 334289DEST_PATH_IMAGE022
(ii) a If the right boundary of the material and the right side edge of the conveying belt have an included angle smaller than a preset value, calculating the gravity center of the material
Figure 197203DEST_PATH_IMAGE023
Distance between projection point on right boundary and right side edge of conveyer belt
Figure 421511DEST_PATH_IMAGE024
And obtaining the first distance.
The invention considers the determining mode of the relative position relation between the conveying belt and the materials under different placing postures of the conveying belt, the placing postures of the materials are rich and diverse when the materials are placed on the conveying belt, and the applicability of the monitoring method of the invention is improved by determining the first interval calculating mode under different postures. When the material is not parallel to the conveying belt, if the minimum distance between the boundary and the side edge meets the threshold condition of the first distance, but the distance between the projection point of the gravity center and the side edge does not meet the threshold condition, the material still has no falling risk according to the stress analysis of the material.
And detecting the first image, and performing image denoising and image enhancement preprocessing by adopting Gaussian filtering and a Laplace operator. The specific operation of gaussian filtering is: each pixel in the image is scanned by a template, and the weighted average gray value of the pixels in the neighborhood determined by the template is used for replacing the value of the central pixel point of the template. The image enhancement by the laplacian operator is realized by that when the gray level of the central pixel of the neighborhood is lower than the average gray level of other pixels in the domain where the central pixel is located, the gray level of the central pixel is further reduced, and when the gray level of the central pixel of the neighborhood is higher than the average gray level of other pixels in the neighborhood where the central pixel is located, the gray level of the central pixel is further improved. Then, a Gaussian scale space is adopted to detect invariant feature points in the scale, and the scale space of the image is expressed into a function
Figure 420691DEST_PATH_IMAGE025
It is a Gaussian function with variable scale
Figure 705042DEST_PATH_IMAGE026
And images
Figure 55251DEST_PATH_IMAGE027
Generated by convolution, i.e.
Figure 83250DEST_PATH_IMAGE028
Wherein
Figure 701051DEST_PATH_IMAGE029
Representing the horizontal and vertical coordinates of the pixels in the image,
Figure 890724DEST_PATH_IMAGE030
representing a gaussian smoothing factor; the method comprises the steps of collecting gradient and direction distribution characteristics of pixels in a Gaussian pyramid image where a key point is located, using the gradient and direction of the pixels in a histogram statistics field, taking the maximum value in the histogram as the main direction of the key point, selecting a plurality of second direction values with the maximum amplitude in the direction of a boundary key point towards the side edge of a conveying belt except the main direction of the key point, obtaining the actual direction value of the key point through interpolation fitting based on the main direction and the second direction values of the key point, connecting the feature point to obtain contour information, identifying the contour information based on a machine learning algorithm to obtain the boundary of a material and the boundary of the conveying belt, and ignoring the boundary information of other foreign matters and conveying belt components on the conveying belt. When the invention determines the direction value of the boundary key point, compared with the prior art that the direction with the amplitude value larger than the preset value is selected from the gradient direction histogram, the invention selects a plurality of direction values with the maximum amplitude value in the direction of the boundary key point facing the side edge of the conveying belt as the auxiliary direction, so that the attribute of the actual direction value of the boundary key point is expanded to the periphery compared with the real boundary of the material key point, the randomness of the direction value of the boundary is reduced, the direction determination speed of the key point is improved, meanwhile, the boundary determined based on the image is larger than or equal to the boundary of the actual material, the prejudgment of the falling state of the material is enhanced, the adjustment of the direction and the speed of the material can be carried out as early as possible, if the material movement is not completely matched with the sampling frequency of a camera, the material can possibly fall between the current sampling period and the next sampling period, the material boundary determination method provided by, in material state identificationThe method realizes the prejudgment effect on the basis, the relative position of the material boundary obtained in the current sampling period and the side edge of the conveying belt is the material state between the current sampling period and the next sampling period, and the defect that the sampling frequency is not completely matched is overcome.
Step S3, judging whether the first distance is smaller than a first distance threshold value, if not, returning to the step S1; if yes, go to step S4.
The first spacing threshold may be preset by a user. As an alternative embodiment, the first distance threshold is determined based on the material size information, in particular if the width information of the material on the conveyor belt is
Figure 728230DEST_PATH_IMAGE031
Then the first pitch threshold may be
Figure 294341DEST_PATH_IMAGE032
。
With the development of the loading and transporting industry, differentiated materials are common material sources in the loading process, and for materials with different sizes, compared with the situation that the falling condition is misjudged easily due to the fact that the same first interval threshold is adopted in the prior art, the first interval threshold is determined according to the material size information, the method can be suitable for application scenes of differential material transmission and loading, and the applicability of the material falling monitoring method is improved.
Further, if the first distance is greater than the first distance threshold, the sampling frequency of the camera is decreased, and the process returns to step S1. Compared with the prior art that materials on the conveying belt are monitored at a fixed frequency, the method and the device have the advantages that the sampling frequency of sampling is determined in a self-adaptive mode according to the state of the materials, the safety of the materials is guaranteed, and meanwhile the sampling frequency is reduced so that energy consumption is reduced.
Step S4, acquiring a second image set of the material, and calculating the movement speed and the movement direction of the material based on the first image and the second image set;
the second image set is a historical image set of the material, the historical image set at least comprises two historical images of the material at the nearest adjacent time, and the moving speed and the moving direction of the material are calculated based on the historical image set and the first image, and the method specifically comprises the following steps:
step S41, determining an effective image area where the first boundary in the first image is located;
step S42 of determining a history image effective region of each history image in the history image set based on the effective image region;
step S43, determining the material boundary in each history image based on the effective region of the history image;
according to the method, the historical image identification area is determined according to the effective image area where the first boundary in the first image is located, when the boundary of the material in the historical image is identified, the boundary is only identified in a smaller area range, and compared with the mode that the boundary is identified in the whole image range in the prior art, the method has the advantages that the effective image area of the historical image is determined based on the effective image area of the first image, the calculation amount of material boundary identification is reduced, and the material state judgment speed is improved.
In step S44, the moving speed and the moving direction of the material are calculated based on the position of the boundary of the material in each history image and the first boundary in the first image.
The history image set contains history images
Figure 533692DEST_PATH_IMAGE001
，
Figure 894266DEST_PATH_IMAGE002
The number of the history images in the history image set is as follows
Figure 219068DEST_PATH_IMAGE003
Calculating the moving speed of two adjacent images according to the time sequence
Figure 588870DEST_PATH_IMAGE004
Wherein, in the step (A),
Figure 918613DEST_PATH_IMAGE005
is a first image
Figure 715668DEST_PATH_IMAGE003
And history image
Figure 262187DEST_PATH_IMAGE006
To get rid of
Figure 435679DEST_PATH_IMAGE002
The speed of the movement of the boundary is,
Figure 384044DEST_PATH_IMAGE007
as a history image
Figure 352000DEST_PATH_IMAGE006
And
Figure 385815DEST_PATH_IMAGE033
to get rid of
Figure 97419DEST_PATH_IMAGE008
The speed of the movement of the boundary is,
Figure 398825DEST_PATH_IMAGE009
as a history image
Figure 537682DEST_PATH_IMAGE010
And
Figure 58794DEST_PATH_IMAGE011
the moving speed of the 1 st boundary between the two, the moving speed of the final material
Figure 246192DEST_PATH_IMAGE012
Wherein, in the step (A),
Figure 231466DEST_PATH_IMAGE013
is shown as
Figure 478908DEST_PATH_IMAGE014
The speed of the movement of the boundary is,
Figure 977061DEST_PATH_IMAGE015
is as follows
Figure 30468DEST_PATH_IMAGE014
The influence factor corresponding to the speed of the boundary motion,
Figure 807931DEST_PATH_IMAGE016
，
Figure 288591DEST_PATH_IMAGE017
。
compared with an average speed calculation mode in the prior art, the method takes the boundary of the material as a reference object and utilizes the influence factor to calculate the comprehensive movement speed of the material, so that on one hand, the movement speed of the material integrates the prior historical speed, and on the other hand, the influence of the movement speed corresponding to the historical image which is closer to the first image in time on the conveying speed of the material is larger during speed synthesis, and the accuracy of calculating the movement speed of the material is improved.
Step S5, calculating the falling probability of the material based on the movement speed, the movement direction and the first distance of the material;
the method comprises the steps of training a first neural network model in advance, wherein input values of the first neural network model are the movement speed, the movement direction and the first distance of a material, and output of the first neural network model is the falling attribute and the probability of the falling attribute of the material, namely the probability corresponding to the falling condition of the material and the probability corresponding to the non-falling condition of the material. And taking the probability corresponding to the falling condition output by the first neural network model as the falling probability of the material.
And step S6, if the falling probability is larger than a first probability threshold, sending out a material falling warning.
If the falling probability is larger than a first probability threshold, identifying material information based on a first image, and sending a material falling warning, wherein the material falling warning comprises material information and prompts monitoring personnel to check and process the material falling condition;
identifying the material information based on the first image includes: detect the information code on the material, the information code can be arbitrary information code such as bar code, two-dimensional code, based on information code discernment material information on the material, and material information includes the brand name, the model etc. of material, sends the material warning that drops, contains material information such as the brand name, the model of material in the material warning that drops, makes monitoring personnel in time know which material has taken place to drop.
As an alternative embodiment, the identifying the material information based on the first image includes: and searching a database based on the material region image in the first image detection to obtain a matched material image and corresponding attribute information thereof, and taking the attribute information as the material information.
As an alternative embodiment, based on the material boundary determining method in step S2, the key points of the material boundary and their main directions are obtained, a first description operator is created for each key point, and the area near the key points is divided into regions in the direction of the material boundary toward the material pixels
Figure 518715DEST_PATH_IMAGE034
And the sub-regions calculate the radius of each sub-region, calculate the characteristic vector of each key point as a first description operator of the key point, and identify the material attribute according to the matching calculation of the first description operator of each key point of the material boundary and the material image in the database. In the invention, because the database only stores the image of the material and does not contain other interference information such as environment, a conveying belt and the like, when the image characteristics of the region where the material is located need to be extracted to be matched with the image in the database, and when the image characteristics of the material region are obtained, compared with the prior art that the sub-regions are divided in the surrounding region of each key point, the method only extracts the image region where the material is located, and only selects the region where the material is located as the sub-region when calculating the first description operator, thereby improving the correlation between the first description operator and the material image and improving the accuracy of material matching identification.
The method and the device only identify the material information when the falling probability is detected to be in the first interval range, and compared with the mode of identifying the falling state after associating the material information with the conveying belt in the prior art, the method and the device only need to identify the information of the material to be fallen, and do not need to associate the state of the normally-transmitted material on the conveying belt with the material information, so that the steps of identifying the material information are reduced, the calculation amount of a monitoring method is reduced, and the monitoring speed is improved.
Further, step S6 further includes:
if the falling probability is smaller than or equal to a first probability threshold and is larger than a second probability threshold, calculating the estimated falling time of the material according to the movement speed, the movement direction and the first distance of the material, and sending a correction signal according to the estimated falling time to correct the movement direction and the movement speed of the material.
When the probability threshold is detected to be in the second interval range, the moving direction and the moving speed of the materials are adjusted by controlling the conveyor belt, so that the materials are prevented from falling. Compared with the prior art that the alarm is carried out once the material leaves the preset position, the invention still processes the material conveying condition in a second interval range in a self-adjusting mode, reduces the frequency of cargo falling alarm, and simultaneously reduces the risk of cargo falling by self-adaptive processing, and reduces the dependence of material conveying on manpower.
Further, after the moving direction and the moving speed of the material are corrected, the sampling frequency of the camera is increased, and the process returns to the step S1.
And if the falling probability is less than or equal to the second probability threshold, reducing the sampling frequency of the camera, and returning to the step S1.
In another preferred embodiment, the sampling frequency f is functionally related to the drop probability p as follows:
Figure 375813DEST_PATH_IMAGE035
；
where f is the sampling frequency, p is the drop probability,
Figure 7782DEST_PATH_IMAGE036
in order to be the basis of the sampling frequency,
Figure 659343DEST_PATH_IMAGE037
the drop probability is based. The relation between the sampling frequency and the falling probability is quantized according to the formula, so that the sampling frequency and the falling probability are well balanced, the human intervention is reduced, and the automation degree is improved.
According to the method, the sampling frequency of the camera is further corrected according to the material state identified in the image, compared with the mode of fixing the sampling frequency of the camera in the prior art, the sampling frequency of the camera is matched with the actual state of the material, the sampling frequency is determined to be increased or decreased based on the real-time material state, the purpose of accurately detecting the falling state of the material is mainly achieved when the material is dangerous, the energy consumption is saved when the material is safe, the accuracy of state detection is considered, and the energy consumption is reduced.
In summary, the embodiment provides a conveying material falling monitoring method based on image recognition, and the material boundary recognition method provided by the invention can be used for prejudging the relative position relationship between the material and the conveying belt on the basis of considering the material recognition accuracy, so that the accuracy of material state monitoring is improved. In addition, the calculation of the material movement speed integrates the movement speed of the previous historical image, and compared with the average value of a plurality of speeds, the accuracy of the movement speed is improved. On the basis of identifying the falling probability of the material, the invention carries out self-adaptive alarm and correction according to the interval range of the falling probability, improves the accuracy of alarm, can reduce the dependence on manpower and automatically corrects the non-emergency condition; in addition, for different material states, the invention can adaptively adjust the sampling frequency of the camera, and gives consideration to the monitoring accuracy and energy consumption of the falling state.
The steps of the above methods are divided for clarity, and the implementation may be combined into one step or split some steps, and the steps are divided into multiple steps, so long as the same logical relationship is included, which are all within the protection scope of the present patent; it is within the scope of the patent to add insignificant modifications to the algorithms or processes or to introduce insignificant design changes to the core design without changing the algorithms or processes.
A third embodiment of the present invention relates to an image recognition-based conveyed material drop monitoring device, including:
the image acquisition module is used for acquiring a first image of the material;
the distance calculation module is used for detecting the first image, identifying a first side edge of the conveying belt and a first boundary of the material, and calculating a first distance between the first side edge and the first boundary;
the judging module is used for judging whether the first distance is smaller than a first distance threshold value;
the movement calculation module is used for acquiring a second image set of the material when the first distance is smaller than a first distance threshold value, and calculating the movement speed and the movement direction of the material based on the first image and the second image set;
the falling probability calculation module is used for calculating the falling probability of the materials based on the movement speed, the movement direction and the first distance of the materials;
and the state processing module is used for sending out a material falling warning if the falling probability is greater than a first probability threshold.
The conveying material falling monitoring device based on image recognition is used for executing instructions of the method in any one of the first embodiment and the second embodiment.
It should be understood that this embodiment is a system example corresponding to the first embodiment, and may be implemented in cooperation with the first embodiment. The related technical details mentioned in the first embodiment are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the first embodiment.
It should be noted that each module referred to in this embodiment is a logical module, and in practical applications, one logical unit may be one physical unit, may be a part of one physical unit, and may be implemented by a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, elements that are not so closely related to solving the technical problems proposed by the present invention are not introduced in the present embodiment, but this does not indicate that other elements are not present in the present embodiment.
The fourth embodiment of the invention provides a conveying material falling monitoring system based on image recognition, which comprises one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing the method of any of embodiments one, two.
Where the memory and processor are connected by a bus, the bus may comprise any number of interconnected buses and bridges, the buses connecting together one or more of the various circuits of the processor and the memory. The bus may also connect various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor is transmitted over a wireless medium via an antenna, which further receives the data and transmits the data to the processor.
The processor is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And the memory may be used to store data used by the processor in performing operations.
It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the invention, and that various changes in form and details may be made therein without departing from the spirit and scope of the invention in practice.
The foregoing description, for purposes of explanation, used specific nomenclature to provide a thorough understanding of the embodiments. It will be apparent, however, to one skilled in the art that the embodiments may be practiced without the specific details. Thus, the foregoing descriptions of specific embodiments described herein are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the embodiments to the precise forms disclosed. It will be apparent to those skilled in the art that many modifications and variations are possible in light of the above teaching. Further, as used herein to refer to the position of a component, the terms above and below, or their synonyms, do not necessarily refer to an absolute position relative to an external reference, but rather to a relative position of the component with reference to the drawings.
Moreover, the foregoing drawings and description include many concepts and features that may be combined in various ways to achieve various benefits and advantages. Thus, features, components, elements and/or concepts from various different figures may be combined to produce embodiments or implementations not necessarily shown or described in this specification. Furthermore, not all features, components, elements and/or concepts shown in a particular figure or description are necessarily required to be in any particular embodiment and/or implementation. It is to be understood that such embodiments and/or implementations fall within the scope of the present description.

Claims (9)

1. A conveying material falling monitoring method based on image recognition is characterized by comprising the following steps:
step S1: collecting a first image of a material;
step S2, detecting the first image, identifying a first side edge of the conveyor belt and a first boundary of the material, and calculating a first distance between the first side edge and the first boundary;
step S3, judging whether the first distance is smaller than a first distance threshold value, if not, returning to the step S1; if yes, go to step S4;
step S4, acquiring a second image set of the material, and calculating the movement speed and the movement direction of the material based on the first image and the second image set;
step S5, calculating the falling probability of the material based on the movement speed, the movement direction and the first distance of the material;
and step S6, if the falling probability is larger than a first probability threshold, sending out a material falling warning.
2. The image recognition-based conveyed material drop monitoring method of claim 1, wherein the step S2 specifically includes:
if the first boundary of the material and the first side edge of the conveying belt are the same, the first distance is the distance between the first boundary of the material and the first side edge of the conveying belt;
if a first included angle exists between the first boundary of the material and the first side edge of the conveying belt, calculating the distance between the projection point of the gravity center of the material on the first boundary and the first side edge of the conveying belt to obtain a first distance.
3. The image recognition-based conveyed material drop monitoring method of claim 1, wherein the step S4 specifically includes:
the second image set is a historical image set of the material,
step S41, determining an effective image area where the first boundary in the first image is located;
step S42 of determining a history image effective region of each history image in the history image set based on the effective image region;
step S43, determining the material boundary in each history image based on the effective region of the history image;
in step S44, the moving speed and the moving direction of the material are calculated based on the position of the boundary of the material in each history image and the first boundary in the first image.
4. Image recognition based conveyed material drop monitoring method according to claim 3,
the history image set contains history images
Figure 83200DEST_PATH_IMAGE001
，
Figure 478409DEST_PATH_IMAGE002
The number of the history images in the history image set is as follows
Figure 964885DEST_PATH_IMAGE003
Calculating the moving speed of two adjacent images according to the time sequence
Figure 52927DEST_PATH_IMAGE004
Wherein, in the step (A),
Figure 257643DEST_PATH_IMAGE005
is a first image
Figure 456543DEST_PATH_IMAGE003
And history image
Figure 296061DEST_PATH_IMAGE006
To get rid of
Figure 492687DEST_PATH_IMAGE002
The speed of the movement of the boundary is,
Figure 247017DEST_PATH_IMAGE007
as a history image
Figure 921712DEST_PATH_IMAGE006
And
Figure 382780DEST_PATH_IMAGE008
to get rid of
Figure 812624DEST_PATH_IMAGE009
The speed of the movement of the boundary is,
Figure 227819DEST_PATH_IMAGE010
as a history image
Figure 768521DEST_PATH_IMAGE011
And
Figure 349675DEST_PATH_IMAGE012
1 st boundary movement speed;
the movement speed of the materials is as follows:
Figure 622525DEST_PATH_IMAGE013
wherein, in the step (A),
Figure 351446DEST_PATH_IMAGE014
is shown as
Figure 367944DEST_PATH_IMAGE015
The speed of the movement of the boundary is,
Figure 302140DEST_PATH_IMAGE016
is as follows
Figure 808207DEST_PATH_IMAGE015
The influence factor corresponding to the speed of the boundary motion,
Figure 962108DEST_PATH_IMAGE017
，
Figure 782297DEST_PATH_IMAGE018
。
5. the image recognition-based conveyed material drop monitoring method of claim 1, wherein the step S6 specifically includes:
if the falling probability is larger than a first probability threshold, identifying material information based on a first image, and sending a material falling warning, wherein the material falling warning comprises material information.
6. The image recognition-based conveyed material drop monitoring method of claim 1, wherein the step S6 specifically includes:
if the falling probability is smaller than or equal to a first probability threshold and is larger than a second probability threshold, calculating the estimated falling time of the material according to the movement speed, the movement direction and the first distance of the material, and sending a correction signal according to the estimated falling time to correct the movement direction and the movement speed of the material.
7. The image recognition-based conveyed material drop monitoring method of claim 6, wherein the step S6 specifically comprises:
after the movement direction and the movement speed of the material are corrected, the sampling frequency of the camera is increased, and the step S1 is returned;
and if the falling probability is less than or equal to the second probability threshold, reducing the sampling frequency of the camera, and returning to the step S1.
8. A conveyor material drop monitoring device based on image recognition is characterized by comprising:
the image acquisition module is used for acquiring a first image of the material;
the distance calculation module is used for detecting the first image, identifying a first side edge of the conveying belt and a first boundary of the material, and calculating a first distance between the first side edge and the first boundary;
the judging module is used for judging whether the first distance is smaller than a first distance threshold value;
the movement calculation module is used for acquiring a second image set of the material when the first distance is smaller than a first distance threshold value, and calculating the movement speed and the movement direction of the material based on the first image and the second image set;
the falling probability calculation module is used for calculating the falling probability of the materials based on the movement speed, the movement direction and the first distance of the materials;
and the state processing module is used for sending out a material falling warning if the falling probability is greater than a first probability threshold.
9. An image recognition based conveyed material drop monitoring system, comprising one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing the method of any of claims 1-7.
CN202110132859.8A 2021-02-01 2021-02-01 Conveying material falling monitoring method, device and system based on image recognition Active CN112446896B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110132859.8A CN112446896B (en) 2021-02-01 2021-02-01 Conveying material falling monitoring method, device and system based on image recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110132859.8A CN112446896B (en) 2021-02-01 2021-02-01 Conveying material falling monitoring method, device and system based on image recognition

Publications (2)

Publication Number Publication Date
CN112446896A true CN112446896A (en) 2021-03-05
CN112446896B CN112446896B (en) 2021-05-11

Family

ID=74740078

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110132859.8A Active CN112446896B (en) 2021-02-01 2021-02-01 Conveying material falling monitoring method, device and system based on image recognition

Country Status (1)

Country Link
CN (1) CN112446896B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113382203A (en) * 2021-05-20 2021-09-10 河南农道智能科技有限公司 AGV goods falling detection method and system based on artificial intelligence
CN115937321A (en) * 2022-09-27 2023-04-07 荣耀终端有限公司 Attitude detection method and device for electronic equipment
CN116311839A (en) * 2023-05-23 2023-06-23 深圳市致尚信息技术有限公司 Set top box drop alarm method and device and electronic equipment
CN116679657A (en) * 2023-08-01 2023-09-01 四川磊蒙机械设备有限公司 Gravel aggregate intelligent monitoring control system
CN117238120A (en) * 2023-11-13 2023-12-15 山西乐宸科技有限公司 Security monitoring method, device, equipment and medium
CN118095995A (en) * 2024-04-26 2024-05-28 运易通科技有限公司 Logistics distribution method and system based on AR technology

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014178936A2 (en) * 2013-04-30 2014-11-06 Gii Acquisition, Llc Dba General Inspection, Llc Method and system for optically inspecting outer peripheral surfaces of parts
CN109051629A (en) * 2018-07-02 2018-12-21 成都精位科技有限公司 The determination method, apparatus and system that material is fallen
CN110047063A (en) * 2019-03-21 2019-07-23 深圳市燕麦科技股份有限公司 A kind of detection method that material is fallen, device, equipment and storage medium
CN111833320A (en) * 2020-07-06 2020-10-27 涵古观智能科技(苏州)有限公司 Method, device and equipment for detecting running state of steel strip and storage medium

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014178936A2 (en) * 2013-04-30 2014-11-06 Gii Acquisition, Llc Dba General Inspection, Llc Method and system for optically inspecting outer peripheral surfaces of parts
CN109051629A (en) * 2018-07-02 2018-12-21 成都精位科技有限公司 The determination method, apparatus and system that material is fallen
CN110047063A (en) * 2019-03-21 2019-07-23 深圳市燕麦科技股份有限公司 A kind of detection method that material is fallen, device, equipment and storage medium
CN111833320A (en) * 2020-07-06 2020-10-27 涵古观智能科技(苏州)有限公司 Method, device and equipment for detecting running state of steel strip and storage medium

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113382203A (en) * 2021-05-20 2021-09-10 河南农道智能科技有限公司 AGV goods falling detection method and system based on artificial intelligence
CN115937321A (en) * 2022-09-27 2023-04-07 荣耀终端有限公司 Attitude detection method and device for electronic equipment
CN115937321B (en) * 2022-09-27 2023-09-22 荣耀终端有限公司 Gesture detection method and device of electronic equipment
CN116311839A (en) * 2023-05-23 2023-06-23 深圳市致尚信息技术有限公司 Set top box drop alarm method and device and electronic equipment
CN116679657A (en) * 2023-08-01 2023-09-01 四川磊蒙机械设备有限公司 Gravel aggregate intelligent monitoring control system
CN116679657B (en) * 2023-08-01 2023-12-01 四川磊蒙机械设备有限公司 Gravel aggregate intelligent monitoring control system
CN117238120A (en) * 2023-11-13 2023-12-15 山西乐宸科技有限公司 Security monitoring method, device, equipment and medium
CN117238120B (en) * 2023-11-13 2024-01-26 山西乐宸科技有限公司 Security monitoring method, device, equipment and medium
CN118095995A (en) * 2024-04-26 2024-05-28 运易通科技有限公司 Logistics distribution method and system based on AR technology

Also Published As

Publication number Publication date
CN112446896B (en) 2021-05-11

Similar Documents

Publication Publication Date Title
CN112446896B (en) Conveying material falling monitoring method, device and system based on image recognition
Aquino et al. A new methodology for estimating the grapevine-berry number per cluster using image analysis
CN104408460B (en) A kind of lane detection and tracking detection method
CN102289660B (en) Method for detecting illegal driving behavior based on hand gesture tracking
EP3176751B1 (en) Information processing device, information processing method, computer-readable recording medium, and inspection system
CN109284674A (en) A kind of method and device of determining lane line
CN109506628A (en) Object distance measuring method under a kind of truck environment based on deep learning
CN111310645A (en) Overflow bin early warning method, device, equipment and storage medium for cargo accumulation amount
CN105373135A (en) Method and system for guiding airplane docking and identifying airplane type based on machine vision
CN103530878A (en) Edge extraction method based on fusion strategy
CN111553310B (en) Security inspection image acquisition method and system based on millimeter wave radar and security inspection equipment
CN111160169A (en) Face detection method, device, equipment and computer readable storage medium
CN111666848B (en) Method, device and equipment for detecting arrival of transport vehicle and storage medium
CN114842560B (en) Computer vision-based construction site personnel dangerous behavior identification method
CN112101260A (en) Method, device, equipment and storage medium for identifying safety belt of operator
CN117876971B (en) Building construction safety monitoring and early warning method based on machine vision
CN116386120A (en) Noninductive monitoring management system
Gal Automatic obstacle detection for USV’s navigation using vision sensors
CN107767366B (en) A kind of transmission line of electricity approximating method and device
CN117197700A (en) Intelligent unmanned inspection contact net defect identification system
CN112347818B (en) Method and device for screening difficult sample images of video target detection model
US6901152B2 (en) Visual profile classification
CN115661800A (en) Dangerous driving behavior detection method based on sight direction time relation learning
CN110177222A (en) A kind of the camera exposure parameter method of adjustment and device of the unused resource of combination vehicle device
CN115496931B (en) Industrial robot health monitoring method and system

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