CN114803386A - Conveying belt longitudinal tearing detection system and method based on binocular line laser camera - Google Patents
Conveying belt longitudinal tearing detection system and method based on binocular line laser camera Download PDFInfo
- Publication number
- CN114803386A CN114803386A CN202210627532.2A CN202210627532A CN114803386A CN 114803386 A CN114803386 A CN 114803386A CN 202210627532 A CN202210627532 A CN 202210627532A CN 114803386 A CN114803386 A CN 114803386A
- Authority
- CN
- China
- Prior art keywords
- damage
- point cloud
- cloud data
- data
- target
- 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
Links
Images
Classifications
-
- 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
- B65G43/02—Control devices, e.g. for safety, warning or fault-correcting detecting dangerous physical condition of load carriers, e.g. for interrupting the drive in the event of overheating
-
- 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
-
- 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
- B65G2203/041—Camera
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Control Of Conveyors (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a system and a method for detecting longitudinal tearing of a conveying belt based on a binocular line laser camera, relating to the technical field of longitudinal tearing detection of the conveying belt, wherein the system comprises the following components: the system comprises a data acquisition module and a back-end processor; the data acquisition module at least comprises a binocular line laser camera; the binocular line laser camera is arranged at the bottom of the target conveyer belt; the binocular line laser camera is used for acquiring point cloud data at the bottom of the target conveyer belt when the target conveyer belt runs; the back-end processor at least comprises a data processing module, and the data processing module is used for determining a feature extraction result corresponding to each frame of point cloud data and a feature extraction result corresponding to the whole target conveying belt when the target conveying belt runs for one week according to the point cloud data at the bottom of the target conveying belt and the transmission speed of the target conveying belt. The invention can achieve the purpose of meeting the detection real-time performance and the detection accuracy.
Description
Technical Field
The invention relates to the technical field of longitudinal tear detection of a conveying belt, in particular to a system and a method for detecting longitudinal tear of a conveying belt based on a binocular line laser camera.
Background
In the coal mine safety production process, the function of the conveying belt is very important. In the coal flow transportation process, foreign matters such as gangue and I-steel possibly scratch the conveying belt, so that huge loss is caused. How to accurately detect the longitudinal tearing problem of the conveying belt is a key object of attention of each coal mine production enterprise.
The existing detection methods for longitudinal tearing of the conveying belt are divided into a contact detection method and a non-contact detection method. Common contact detection methods include a force measurement method, an embedding method, a coil detection method, a tension detection method and the like; non-contact detection methods include closed coil methods, X-ray methods, machine vision methods, and the like. The contact detection method is gradually replaced by a non-contact detection method due to low accuracy and abrasion. The machine vision method in the non-contact detection method gets more and more attention and research due to the advantages of small loss, high accuracy, simple maintenance and the like.
The detection method for longitudinal tearing of the conveying belt based on machine vision mainly comprises two types: the first method comprises the following steps: for the analysis of visible or infrared light images, the second category of methods: the analysis is performed on the linear assist laser image. The first method has the key point that the tearing damage on the conveying belt image is accurately segmented by adopting a maximum inter-class variance method, a threshold iteration method and a global threshold method. The second method judges whether the conveyor belt has tearing damage or not by means of the characteristic change of linear laser stripes projected on the surface of the conveyor belt, and specifically comprises the following steps: the method comprises the steps of projecting a single linear laser on the surface of a conveying belt, extracting a light strip framework by using a maximum value method, determining a breakpoint position by using a neighborhood difference, judging a fluctuation abnormal position by using a second derivative, further detecting and marking a longitudinal tearing area of the conveying belt, and identifying longitudinal tearing by detecting fracture characteristics of a single linear laser contour line projected on the surface of the conveying belt.
When the first method is applied to underground coal mines, due to the fact that light is weak and uneven, dust and humidity are large, the conveying belt vibrates up and down in the running process and the surface abrasion degree is different, the definition of collected images is low and even fuzzy, and tearing damage and segmentation are difficult. When judging whether the tearing damage exists, the second method only focuses on whether the longitudinal tearing exists or not, does not relate to calculation of characteristic information such as length, width and depth of the tearing damage, or cannot evaluate the characteristic information such as the length, width and depth of the tearing damage with high precision, and can evaluate the damage position only by auxiliary means such as extra markers.
Disclosure of Invention
The invention aims to provide a conveying belt longitudinal tearing detection system and method based on a binocular line laser camera, and the purpose of meeting detection real-time performance and detection accuracy is achieved.
In order to achieve the purpose, the invention provides the following scheme:
the utility model provides a conveyer belt is vertical tears detecting system based on binocular line laser camera, includes: the system comprises a data acquisition module and a back-end processor;
the data acquisition module at least comprises a binocular line laser camera; the binocular line laser camera is arranged at the bottom of the target conveyer belt; the binocular line laser camera is used for acquiring point cloud data at the bottom of the target conveyer belt when the target conveyer belt runs;
the back-end processor comprises at least one data processing module, and the data processing module is used for:
acquiring point cloud data of the bottom of the target conveyer belt;
carrying out damage characteristic information extraction on the point cloud data, and determining a characteristic extraction result corresponding to each frame of point cloud data; the feature extraction result comprises a non-damage feature result and a damage feature result; the lesion signature results include the location, category, and size of the lesion; the categories comprise tearing crack damage of the conveying belt and tearing deformation, distortion and folding damage of the conveying belt; the dimensions include one or more of length, width, depth;
and fusing the feature extraction results corresponding to the multi-frame point cloud data based on the transmission speed of the target conveyer belt so as to determine the feature extraction result of the whole target conveyer belt corresponding to the target conveyer belt running for one week.
Optionally, the point cloud data is distributed in a linear array form; before performing damage feature information extraction on the point cloud data and determining the feature extraction result corresponding to each frame of the point cloud data, the data processing module is further configured to:
and preprocessing the point cloud data by adopting a statistical outlier removal algorithm.
Optionally, in terms of performing damage feature information extraction on the point cloud data and determining a feature extraction result corresponding to each frame of the point cloud data, the data processing module is further configured to perform damage feature information extraction on the point cloud data, and determine a feature extraction result corresponding to each frame of the point cloud data
Judging whether the preprocessed point cloud data is damaged or not, if so, determining that a feature extraction result corresponding to the preprocessed point cloud data is a non-damaged feature result, and if not, determining a damage type and a damage area according to the preprocessed point cloud data;
when the damage type is the damage of the tearing crack of the conveying belt, calculating the damage length and the damage width according to the damage area;
and when the damage type is the tearing, deformation, distortion and folding damage of the conveying belt, calculating the damage length, the damage width and the damage depth according to the damage area.
Optionally, the data processing module is further configured to, after determining whether the preprocessed point cloud data is damaged, if so, determine that a feature extraction result corresponding to the preprocessed point cloud data is a non-damaged feature result, and if not, determine a damage category and a damage region according to the preprocessed point cloud data, where:
and judging that the distance between any adjacent data points in the preprocessed point cloud data is smaller than a first threshold and the slope between any adjacent data points in the preprocessed point cloud data is smaller than a second threshold, if so, determining that the feature extraction result corresponding to the preprocessed point cloud data is a non-damage feature result, and otherwise, determining the damage category and the damage area according to the preprocessed point cloud data.
Optionally, in determining the damage category and the damage region according to the preprocessed point cloud data, the data processing module is further configured to:
when the distance between adjacent data points in the preprocessed point cloud data is larger than or equal to a first threshold value and the slope between any adjacent data points in the preprocessed point cloud data is smaller than a second threshold value, determining that the preprocessed point cloud data is damaged, determining the damage type corresponding to the preprocessed point cloud data as the damage of the tearing crack of the conveyor belt, and determining the region surrounded by the first target data points as the damage region corresponding to the preprocessed point cloud data; the first target data point satisfies that a distance between the first target data point and a data point adjacent to the first target data point is greater than or equal to a first threshold;
when the distance between any adjacent data points in the preprocessed point cloud data is smaller than a first threshold and the slope between the adjacent data points in the preprocessed point cloud data is larger than or equal to a second threshold, determining that the preprocessed point cloud data is damaged, determining the damage type corresponding to the preprocessed point cloud data as the tearing, deforming, distorting and folding damage of the conveyor belt, and determining the area surrounded by the second target data points as the damaged area corresponding to the preprocessed point cloud data; the second target data point satisfies that a slope between the second target data point and a data point adjacent to the second target data point is greater than or equal to a second threshold.
Optionally, when the damage category is a tear crack damage of the conveyor belt, the data processing module is further configured to calculate a damage length and a damage width according to the damage region
When the damage category is the damage of the tearing crack of the conveying belt, according to a formula
Calculating the damage length; wherein Dis1 is the actual damage length, len1 is the damage length calculated by the preprocessed point cloud data, V is the transmission speed of the target conveyor belt, t is the single-frame data time, and num is the point cloud column number of the single-frame preprocessed point cloud data;
when the damage category is the damage of the tearing crack of the conveying belt, according to a formula
Calculating the width of the damage; where Dis2 is the actual lesion width and len2 is the lesion width calculated from the preprocessed point cloud data.
Optionally, the method further comprises: a back end control unit;
the input end of the rear-end control unit is connected with the output end of the data processing module, and the output end of the rear-end control unit is connected with the audible and visual alarm;
the back end control unit is configured to:
acquiring a feature extraction result corresponding to each frame of point cloud data and a feature extraction result of the whole target conveyor belt;
respectively comparing a feature extraction result corresponding to each frame of point cloud data and a feature extraction result of the whole target conveyor belt with alarm constraint conditions, and outputting an acousto-optic alarm instruction when the feature extraction result corresponding to the point cloud data and/or the feature extraction result of the whole target conveyor belt meet any one of the alarm constraint conditions; the alarm constraint conditions comprise three constraint conditions, namely a damage length threshold, a damage width threshold and a damage depth threshold.
Optionally, the method further comprises: an audible and visual alarm;
and the audible and visual alarm is used for executing audible and visual alarm operation according to the received audible and visual alarm instruction.
Optionally, the binocular line laser camera is installed at the bottom of the target conveyer belt in an obliquely upward 45 ° installation manner.
A method for detecting longitudinal tearing of a conveying belt based on a binocular line laser camera comprises the following steps:
acquiring point cloud data of the bottom of a target conveyer belt; the point cloud data is collected by a binocular line laser camera arranged at the bottom of the target conveyer belt;
carrying out damage characteristic information extraction on the point cloud data, and determining a characteristic extraction result corresponding to each frame of point cloud data; the feature extraction result comprises a non-damage feature result and a damage feature result; the lesion signature results include the location, category, and size of the lesion; the categories comprise tearing crack damage of the conveying belt and tearing deformation, distortion and folding damage of the conveying belt; the dimensions include one or more of length, width, depth;
and fusing the feature extraction results corresponding to the multi-frame point cloud data based on the transmission speed of the target conveyer belt so as to determine the feature extraction result of the whole target conveyer belt corresponding to the target conveyer belt running for one week.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a conveyor belt longitudinal tearing detection system and method based on a binocular line laser camera. Firstly, identifying point cloud characteristics of binocular line laser projected at the bottom of a target conveyer belt in real time on line through a point cloud processing technology; secondly, determining multi-frame point cloud characteristics required during fusion based on data collected by a speed sensor arranged on a belt conveyor; and then fusing the required multi-frame point cloud characteristics based on a fusion technology to realize the identification of the longitudinal tearing characteristics of the whole target conveying belt, and further carrying out integrated quantitative analysis on the width, depth and length of the damage. Compared with single-channel linear laser and multi-channel linear laser, the dual-eye line laser adopted by the invention improves the damage detection precision and reduces the estimation error of the damage length.
The method is low in complexity, can be operated in field analysis and decision of the mine intrinsically safe embedded equipment, does not need to transmit data to an upper computer, avoids time delay, and meets the requirements of real-time performance and accuracy of detection.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
Fig. 1 is a schematic view of a shooting range of a binocular line laser camera according to an embodiment of the present invention;
fig. 2 is a structural diagram of a conveyor belt longitudinal tear detection system based on a binocular line laser camera according to an embodiment of the present invention;
FIG. 3 is a diagram of a conveyor belt tear seam provided in accordance with an embodiment of the present invention;
FIG. 4 is a schematic view of a conveyor belt tearing, deforming, twisting and folding apparatus according to an embodiment of the present invention;
FIG. 5 is an expanded view of a crack detection plane provided by an embodiment of the present invention;
FIG. 6 is a flowchart illustrating the operation of a binocular laser camera based conveyor belt longitudinal tear detection system according to an embodiment of the present invention;
fig. 7 is a flowchart of a method for detecting longitudinal tear of a conveyor belt based on a binocular line laser camera according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and 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.
The invention aims to provide a conveyor belt longitudinal tearing detection system and method based on a binocular line laser camera, which mainly detect the longitudinal tearing of the conveyor belt by using a binocular ranging technology, a point cloud processing technology and the like. The method specifically comprises the following steps: the method comprises the steps of acquiring point cloud data at the bottom of the conveying belt in real time by using a binocular line laser camera, processing single-frame point cloud data to acquire related damage characteristic information of the conveying belt, and finishing calculation of longitudinal tearing damage size and alarm signal output of the whole conveying belt according to a multi-frame point cloud data fusion processing method. The invention can carry out damage defect detection and damage size calculation on the conveying belt in real time and output alarm information, and has great application value for intelligent detection and protection of the belt conveyor.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Example one
The method comprises the steps of marking positions, such as a transfer point of a conveying belt, a machine head and the like, which are easy to longitudinally tear, installing a binocular line laser camera below the marking position of the conveying belt, installing the binocular line laser camera at an angle of 45 degrees upwards to shoot the bottom of the conveying belt, enabling the shooting range of the binocular line laser camera to be as shown in figure 1, utilizing point cloud data at the bottom of the conveying belt to extract and analyze point cloud characteristics, completing longitudinal tearing detection according to the crack defect characteristics and the folding defect characteristics of the conveying belt, and finally performing alarm judgment according to counted continuous multiframe damage information.
As shown in fig. 2, the conveyor belt longitudinal tear detection system based on the binocular line laser camera provided by the embodiment of the invention mainly comprises a data acquisition module and a back-end processor.
The data acquisition module at least comprises a binocular line laser camera; the binocular line laser camera is arranged at the bottom of the target conveyer belt; the binocular line laser camera is used for acquiring point cloud data at the bottom of the target conveyer belt when the target conveyer belt runs; wherein the target conveyor belt is a conveyor belt to be measured. The binocular line laser camera operates at a fixed frequency.
Preferably, the data acquisition module further comprises a protective shell for protecting the binocular line laser camera.
The back-end processor comprises at least one data processing module. The data processing module is used for:
firstly, point cloud data of the bottom of the target conveyor belt are obtained; secondly, performing damage characteristic information extraction on the point cloud data, and determining a characteristic extraction result corresponding to each frame of point cloud data; the feature extraction result comprises a non-damage feature result and a damage feature result; the lesion signature results include the location, category, and size of the lesion; the categories comprise tearing crack damage of the conveying belt and tearing deformation, distortion and folding damage of the conveying belt; the dimensions include one or more of length, width, depth; and finally, fusing the feature extraction results corresponding to the multi-frame point cloud data based on the transmission speed of the target conveyer belt so as to determine the feature extraction result of the whole target conveyer belt corresponding to the target conveyer belt running for one week.
The hardware structure of the data processing module is an RV1126 development board.
Further, the back-end processor further includes an auxiliary analog quantity input module (the auxiliary analog quantity input module is used for acquiring analog quantity signals output by other platforms/devices, and then loading the analog quantity signals into the data processing module for algorithm simulation), an auxiliary analog quantity output module (the auxiliary analog quantity output module is used for outputting a processing result output by the data processing module as an analog quantity, and then transmitting the analog quantity to an upper computer for alarm or target conveyor belt control through a modbustcp protocol), a DC24V input module (the DC24V input module is used for providing 24V direct current for the RV1126 development board), a DC24V output module (the RV1126 development board outputs 24V direct current for the expansion board through the DC24V output module), and a system management module (the system management module is used for setting parameters of the back-end processor).
And a related application algorithm and an interactive interface are transplanted in the data processing module to complete the extraction of the related characteristics of the point cloud data, the longitudinal tearing detection processing and the expression of the algorithm result. The algorithm result expression comprises analog quantity output and is compatible with common network communication output such as Modbustcp, TCP, database and the like.
According to the conveyor belt longitudinal tearing detection system based on the binocular line laser camera, the binocular line laser camera is used for acquiring the point cloud data at the bottom of the conveyor belt in real time, and longitudinal tearing damage detection and multi-frame information fusion calculation are carried out on each frame of point cloud data.
As a preferred implementation manner, the determining process of the feature extraction result according to the embodiment of the present invention is:
the collected conveyer belt point cloud data is from the bottom of the conveyer belt, so the obtained point cloud data is distributed in a linear array form, and the point cloud data to be processed in the embodiment of the invention is the point cloud data based on rules.
Due to the problem of the field range of the laser device, it may happen that the field of view is large and devices other than the conveyor belt device are extracted into the point cloud data, so that the point cloud data needs to be preprocessed before feature extraction, mainly in order to remove the isolated local point cloud data of the group, so that the processed point cloud data only contains the point cloud data describing the conveyor belt device. The embodiment of the invention mainly uses a statistical outlier removing method to remove outlier isolated local point cloud data, and specifically uses a PCL point cloud database static _ outlier _ removal function to set the number of field points and standard deviation parameters to realize the functions.
Before performing damage feature information extraction on the point cloud data and determining a feature extraction result corresponding to each frame of the point cloud data, the data processing module is further configured to:
and preprocessing the point cloud data by adopting a statistical outlier removal algorithm.
The tearing damage of the conveying belt is mainly shown in two conditions, one is tearing cracks of the conveying belt caused by scratching and the like, and the other is tearing, deforming, twisting and folding of the conveying belt, which is particularly shown in fig. 3 and 4.
(1) Conveyor belt tear crack damage detection
The tearing crack damage of the conveying belt is mainly represented in the structural characteristics of point cloud data, namely the distance between adjacent data points of the tearing crack damage is larger than that between normal adjacent points.
In order to more efficiently realize the damage detection of the conveying belt cracks, the crack damage is insensitive to the point cloud depth information, so that the point cloud data depth information is removed, and only the two-dimensional plane information is reserved. The points are arranged in rows and at intervals according to the distance of the point clouds in each row, as shown in fig. 5.
The main principle of crack detection is as follows: and calculating the distance between every two adjacent points, setting a crack detection threshold value, marking the point as a crack point if the distance between the adjacent points is greater than or equal to the crack detection threshold value, and connecting the found crack points in sequence to obtain the whole crack area. The size of the crack detection threshold directly affects the crack detection accuracy, so the crack detection threshold needs to be set reasonably. The size of the crack detection threshold provided by the embodiment of the invention mainly depends on the density degree of the three-dimensional point cloud data and the average value of the distances between adjacent points. Selecting an average value of distances between all adjacent rows of points which are generally k times of the crack detection threshold, wherein k is an adjusting coefficient and is more than or equal to 1; the method and the device are selected according to the density degree of the three-dimensional point cloud data, the smaller the k value is, the smaller the detected crack is, and k is 1.3 selected in the embodiment of the invention.
Calculating the crack detection damage degree:
since the interval distance of each row of point clouds and the interval distance of each row of point clouds are known during the acquisition of point cloud data on the surface of the conveyor belt, in the calculation of the crack damage degree, the crack width is equal to the average width of all adjacent crack points on the upper and lower boundaries, as shown in fig. 5, AD and BC are the widths of two boundaries, and the crack length is equal to the distance from the starting crack boundary point to the ending crack boundary point, that is, the product of the interval distance h of each row of point clouds and the row number N of point clouds in fig. 5. And recording the coordinate information of the initial boundary point and the coordinate information of the final boundary point of the crack detection area, and providing subsequent multi-frame information fusion processing. The actual distance length needs to be calculated from the current conveyor speed data, i.e.:
where Dis is the actual distance length, len is the calculated length of the point cloud data, V is the conveyor belt speed, t is the single frame data time, and num is the number of point cloud columns, i.e., the number of point clouds, of the single frame point cloud data.
(2) Detection of tearing, deformation, distortion, folding and damage of conveying belt
The tearing, deformation, distortion and folding damage of the conveying belt is mainly shown as that the conveying belt has obvious corners in a point cloud data characteristic structure, the slope change between adjacent points of the local point cloud is severe, and the processes of rising and falling of the slope between the adjacent point clouds are obvious. Based on the damage characteristic, the embodiment of the invention processes the three-dimensional point cloud data of the conveying belt, realizes the tearing, deformation, distortion and folding damage of the conveying belt, and completes the calculation of the damage degree.
The main principle of tearing, deforming, distorting and folding damage of the conveying belt is as follows: calculating the slope between each group of adjacent points, setting a detection threshold of the damage type, marking the point as a boundary point of the damage type if the slope change of the adjacent points is larger than the preset detection threshold, and connecting the found boundary points of the types in sequence to obtain the whole area of the damage type. The size of the damage detection threshold directly influences the precision of tearing, deforming, distorting, folding and damaging the conveying belt, so that the size of the detection threshold needs to be set reasonably. The size of the detection threshold value of the tearing, deformation, distortion and folding damage of the conveying belt in the embodiment of the invention mainly depends on the density degree of three-dimensional point cloud data and the slope value of adjacent point data. The invention selects the threshold value as dynamic change, namely the slope value between the current two adjacent points can not be larger than 2 times of the slope value between the previous two adjacent point clouds, otherwise, the threshold value is regarded as a damaged area point.
Calculating the tearing, deformation, distortion and folding damage degree of the conveying belt: and extracting the detected damage areas of the type to perform cube fitting, obtaining an envelope cube which envelopes the damage point cloud areas at the minimum, and recording the coordinates, the length, the width and the height of the outermost periphery boundary point of the cube. The width is the width of the type of damage, the height is the depth of the type of damage, and the length is the length of a single frame of the type of damage. And the coordinate information of the outermost boundary point of the cube provides subsequent multi-frame information fusion processing. The actual distance length needs to be calculated from the current conveyor speed data, i.e.:
dis is the actual distance length, len is the calculated length of the point cloud data, V is the conveyor belt speed, t is the single frame data time, and num is the number of point cloud columns of the single frame point cloud data, i.e. the number of point cloud strips.
Therefore, in terms of extracting damage feature information of the point cloud data and determining a feature extraction result corresponding to each frame of the point cloud data, the data processing module is further configured to:
judging whether the preprocessed point cloud data is damaged or not, if so, determining that a feature extraction result corresponding to the preprocessed point cloud data is a non-damaged feature result, and if not, determining the damage type and the damage area according to the preprocessed point cloud data.
And when the damage type is the damage of the tearing crack of the conveying belt, calculating the damage length and the damage width according to the damage area.
And when the damage type is the tearing, deformation, distortion and folding damage of the conveying belt, calculating the damage length, the damage width and the damage depth according to the damage area.
Further, in the aspect of judging whether the preprocessed point cloud data is damaged or not, if so, determining that a feature extraction result corresponding to the preprocessed point cloud data is a non-damaged feature result, and if not, determining a damage category and a damage area according to the preprocessed point cloud data, the data processing module is further configured to:
and judging that the distance between any adjacent data points in the preprocessed point cloud data is smaller than a first threshold and the slope between any adjacent data points in the preprocessed point cloud data is smaller than a second threshold, if so, determining that the feature extraction result corresponding to the preprocessed point cloud data is a non-damage feature result, and otherwise, determining the damage category and the damage area according to the preprocessed point cloud data.
Further, in terms of determining the damage category and the damage area according to the preprocessed point cloud data, the data processing module is further configured to:
when the distance between adjacent data points in the preprocessed point cloud data is larger than or equal to a first threshold value and the slope between any adjacent data points in the preprocessed point cloud data is smaller than a second threshold value, determining that the preprocessed point cloud data is damaged, determining the damage type corresponding to the preprocessed point cloud data as the damage of the tearing crack of the conveyor belt, and determining the region surrounded by the first target data points as the damage region corresponding to the preprocessed point cloud data; the first target data point satisfies that a distance between the first target data point and a data point adjacent to the first target data point is greater than or equal to a first threshold.
When the distance between any adjacent data points in the preprocessed point cloud data is smaller than a first threshold and the slope between the adjacent data points in the preprocessed point cloud data is larger than or equal to a second threshold, determining that the preprocessed point cloud data is damaged, determining the damage type corresponding to the preprocessed point cloud data as the tearing, deforming, distorting and folding damage of the conveyor belt, and determining the area surrounded by the second target data points as the damaged area corresponding to the preprocessed point cloud data; the second target data point satisfies that a slope between the second target data point and a data point adjacent to the second target data point is greater than or equal to a second threshold.
When the damage category is the damage of the tearing crack of the conveying belt, the data processing module is also used for calculating the damage length and the damage width according to the damage area
When the damage category is the damage of the tearing crack of the conveying belt, according to a formula
Calculating the damage length; where Dis1 is the actual damage length, len1 is the damage length calculated by the preprocessed point cloud data, V is the transmission speed of the target conveyor belt, t is the single-frame data time, and num is the number of point cloud columns of the single-frame preprocessed point cloud data.
When the damage category is the damage of the tearing crack of the conveying belt, according to a formula
Calculating the width of the damage; where Dis2 is the actual lesion width and len2 is the lesion width calculated from the preprocessed point cloud data.
The same reason is according to
And (4) performing deduction, and calculating the damage length, the damage width and the damage depth according to the damaged area when the damage type is the tearing, deformation, distortion and folding damage of the conveying belt.
As a preferred implementation manner, the fusion process described in the embodiment of the present invention is:
when the related type damage end point is detected, the partial damage detection and recording are finished. In order to obtain the characteristic values (including length, width and depth information) of the whole complete damage, multi-frame point cloud data from a damage starting frame to an injury ending frame needs to be analyzed.
When the damage detection end points of the single-frame point cloud data are distributed at the tail part of the frame point cloud data, monitoring the point cloud data of the next frame, and when the start position of the damage area detected by the latest frame point cloud data is superposed with the end position detected by the last frame point cloud data (the distance between the start position and the end position is not more than 1.5 times of the distance between the two row point clouds), merging the damage areas of the two frames, including length merging, width maximum value taking and depth maximum value taking.
(3) Alarm output detection
And aiming at the feature extraction result corresponding to each frame of point cloud data and the feature extraction result of the whole target conveyor belt, judging whether the longitudinal tearing condition exceeding the alarm limit threshold value occurs or not in real time, at the moment, outputting application logic according to the judged result by an algorithm, transmitting an analog signal to a rear-end control unit PLC, and carrying out subsequent linkage control on the PLC according to the received signal, wherein the subsequent linkage control comprises conveyor belt shutdown control, acousto-optic alarm control and the like. The alarm limit threshold value can set the relevant threshold value to be parameter configurable according to expert prior knowledge and system flexibility.
The general case configuration is as follows:
according to the speed of 4.5m/s of the conveyer belt, the damage with the damage length of more than 20m, the damage with the width of more than 5mm and the damage with the depth of 5-20mm is subjected to sound-light alarm and relevant operation (the conveyer belt stops running and the like).
Based on this, the conveyor belt longitudinal tearing detection system based on the binocular line laser camera in the embodiment of the present invention further includes: the rear end control unit and the acousto-optic alarm; the input end of the rear-end control unit is connected with the output end of the data processing module, and the output end of the rear-end control unit is connected with the audible and visual alarm.
The back end control unit is used for: acquiring a feature extraction result corresponding to each frame of point cloud data and a feature extraction result of the whole target conveyor belt, respectively comparing the feature extraction result corresponding to each frame of point cloud data and the feature extraction result of the whole target conveyor belt with alarm constraint conditions, and outputting an acousto-optic alarm instruction when the feature extraction result corresponding to the point cloud data and/or the feature extraction result of the whole target conveyor belt meet any one of the alarm constraint conditions; the alarm constraint conditions comprise three constraint conditions, namely a damage length threshold, a damage width threshold and a damage depth threshold.
And the audible and visual alarm is used for executing audible and visual alarm operation according to the received audible and visual alarm instruction.
Preferably, the back end control unit is a PLC or other devices.
Analog quantity signals output by the back-end processor are transmitted to a back-end control unit (PLC) (in a mode of Modbustcp communication protocol, corresponding values are input at the back end, and are transmitted to the PLC through the protocol), and the PLC carries out linkage control or alarm output. The linkage control comprises the shutdown control and the audible and visual alarm control of the conveying belt, once the algorithm detects that the longitudinal tearing phenomenon of the conveying belt occurs and reaches the related set alarm limit, the audible and visual alarm is immediately carried out, the shutdown operation is carried out on the conveying belt, and the larger loss is avoided.
Further, the back-end processor is connected with the back-end control unit through an electric cable or an optical cable.
As a preferred implementation manner, the conveyor belt longitudinal tear detection system based on the binocular line laser camera according to the embodiment of the present invention further includes: a control and display part (comprising various control communication interfaces, management platforms, executors, decoders and the like), and a data storage part (comprising servers, magnetic disks and the like).
Data storage and optimization
The historical point cloud data and the like of longitudinal tearing damage at the bottom of the conveying belt are stored in the server, historical alarm records can be checked to form a report, and algorithm parameters are continuously optimized and adjusted through data analysis.
The data processing flow of the conveyor belt longitudinal tearing detection system based on the binocular line laser camera is shown in fig. 6, and specifically comprises the following steps:
point cloud data acquisition- > point cloud data processing (extraction of conveyer belt damage characteristic information) — conveyer belt damage defect detection- > multi-frame data fusion calculation- > alarm threshold judgment- > application logic output (back end control unit).
As shown in fig. 6, the system proposed by the embodiment of the present invention mainly includes two parts, namely data analysis and multi-frame information fusion. The data analysis is mainly used for processing the single-frame laser point cloud data of the conveyer belt so as to extract the tearing damage characteristics of the single-frame point cloud data of the conveyer belt and the point cloud position, width and depth information of the single-frame point cloud data of the conveyer belt; the multi-frame information fusion is mainly used for carrying out fusion calculation on the detection result of the tearing damage of continuous multi-frame point cloud data of the conveying belt, so that the damage information (including damage length, width and depth) of the integral continuous multi-frame conveying belt is counted through the information, finally, the counted damage information is compared with a conveying belt longitudinal tearing alarm threshold value set by expert experience, and when the damage information exceeds the set alarm threshold value, sound-light alarm is carried out and corresponding appointed action is carried out.
Example two
In order to achieve the above object, the present invention further provides a method for detecting longitudinal tear of a conveyor belt based on a binocular line laser camera, as shown in fig. 7, the method includes:
step 100: acquiring point cloud data of the bottom of a target conveyer belt; the point cloud data is collected by a binocular line laser camera arranged at the bottom of the target conveyer belt;
step 200: carrying out damage characteristic information extraction on the point cloud data, and determining a characteristic extraction result corresponding to each frame of point cloud data; the feature extraction result comprises a non-damage feature result and a damage feature result; the lesion signature results include the location, category, and size of the lesion; the categories comprise tearing crack damage of the conveying belt and tearing deformation, distortion and folding damage of the conveying belt; the dimensions include one or more of length, width, depth;
step 300: and fusing the feature extraction results corresponding to the multi-frame point cloud data based on the transmission speed of the target conveyer belt so as to determine the feature extraction result of the whole target conveyer belt corresponding to the target conveyer belt running for one week.
Compared with the prior art, the invention has the following advantages:
the laser equipment can set the density and the quantity of the collected point clouds according to the precision and the timeliness, and can realize the longitudinal tearing detection of the conveying belt by matching with the idea of concurrent blocking treatment, the detection speed can be realized within 500ms, the accuracy rate reaches 99 percent, and the false alarm rate is less than 1 percent.
The core algorithm is realized and deployed on a board of a side-end device RV1126, the device is a mining intrinsic safety type device, the system stability is good, the fault tolerance rate is high, and the device is not interfered by the electromagnetic environment of the environment; the embedded algorithm is transplanted, edge calculation is supported, the operation pressure of the central server can be greatly reduced, the underground looped network still keeps normal work when the underground looped network has problems, and the stability of the system is improved.
The system longitudinally tears warning threshold value is nimble to be joined in marriage, can carry out the interactive use with any other equipment, and system compatibility is strong, covers many scenes of polymorphic type conveyer belt and tears detection longitudinally.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (10)
1. The utility model provides a vertical detecting system that tears of conveyer belt based on binocular line laser camera which characterized in that includes: the system comprises a data acquisition module and a back-end processor;
the data acquisition module at least comprises a binocular line laser camera; the binocular line laser camera is arranged at the bottom of the target conveyer belt; the binocular line laser camera is used for acquiring point cloud data at the bottom of the target conveyer belt when the target conveyer belt runs;
the back-end processor comprises at least one data processing module, and the data processing module is used for:
acquiring point cloud data of the bottom of the target conveyer belt;
performing damage characteristic information extraction on the point cloud data, and determining a characteristic extraction result corresponding to each frame of point cloud data; the feature extraction result comprises a non-damage feature result and a damage feature result; the lesion signature results include the location, category, and size of the lesion; the categories comprise tearing crack damage of the conveying belt and tearing deformation, distortion and folding damage of the conveying belt; the dimensions include one or more of length, width, depth;
and fusing the feature extraction results corresponding to the multi-frame point cloud data based on the transmission speed of the target conveyer belt so as to determine the feature extraction result of the whole target conveyer belt corresponding to the target conveyer belt running for one week.
2. The binocular line laser camera based conveyor belt longitudinal tear detection system of claim 1, wherein the point cloud data is distributed in a linear array; before performing damage feature information extraction on the point cloud data and determining the feature extraction result corresponding to each frame of the point cloud data, the data processing module is further configured to:
and preprocessing the point cloud data by adopting a statistical outlier removal algorithm.
3. The binocular line laser camera-based conveyor belt longitudinal tear detection system of claim 2, wherein the data processing module is further configured to extract damage feature information of the point cloud data and determine a feature extraction result corresponding to each frame of the point cloud data
Judging whether the preprocessed point cloud data is damaged or not, if so, determining that a feature extraction result corresponding to the preprocessed point cloud data is a non-damaged feature result, and if not, determining a damage type and a damage area according to the preprocessed point cloud data;
when the damage type is the damage of the tearing crack of the conveying belt, calculating the damage length and the damage width according to the damage area;
and when the damage type is the tearing, deformation, distortion and folding damage of the conveying belt, calculating the damage length, the damage width and the damage depth according to the damage area.
4. The binocular line laser camera-based conveyor belt longitudinal tear detection system of claim 3, wherein in determining whether the preprocessed point cloud data is not damaged, if so, determining that a feature extraction result corresponding to the preprocessed point cloud data is a non-damaged feature result, otherwise, determining a damage category and a damage area according to the preprocessed point cloud data, the data processing module is further configured to:
and judging that the distance between any adjacent data points in the preprocessed point cloud data is smaller than a first threshold and the slope between any adjacent data points in the preprocessed point cloud data is smaller than a second threshold, if so, determining that the feature extraction result corresponding to the preprocessed point cloud data is a non-damage feature result, and otherwise, determining the damage category and the damage area according to the preprocessed point cloud data.
5. The binocular line laser camera-based conveyor belt longitudinal tear detection system of claim 3, wherein in determining the damage category and the damage region from the pre-processed point cloud data, the data processing module is further configured to:
when the distance between adjacent data points in the preprocessed point cloud data is larger than or equal to a first threshold value and the slope between any adjacent data points in the preprocessed point cloud data is smaller than a second threshold value, determining that the preprocessed point cloud data is damaged, determining the damage type corresponding to the preprocessed point cloud data as the damage of the tearing crack of the conveyor belt, and determining the region surrounded by the first target data points as the damage region corresponding to the preprocessed point cloud data; the first target data point satisfies that a distance between the first target data point and a data point adjacent to the first target data point is greater than or equal to a first threshold;
when the distance between any adjacent data points in the preprocessed point cloud data is smaller than a first threshold and the slope between the adjacent data points in the preprocessed point cloud data is larger than or equal to a second threshold, determining that the preprocessed point cloud data is damaged, determining the damage type corresponding to the preprocessed point cloud data as the tearing, deforming, distorting and folding damage of the conveyor belt, and determining the area surrounded by the second target data points as the damaged area corresponding to the preprocessed point cloud data; the second target data point satisfies that a slope between the second target data point and a data point adjacent to the second target data point is greater than or equal to a second threshold.
6. The binocular line laser camera-based conveyor belt longitudinal tear detection system of claim 3, wherein the data processing module is further configured to calculate a length of the tear and a width of the tear according to the damaged area when the damage category is a conveyor belt tear crack damage
When the damage category is the damage of the tearing crack of the conveying belt, according to a formula
Calculating the length of the damage; wherein Dis1 is the actual damage length, len1 is the damage length calculated by the preprocessed point cloud data, V is the transmission speed of the target conveyor belt, t is the single-frame data time, and num is the point cloud column number of the single-frame preprocessed point cloud data;
when the damage category is the damage of the tearing crack of the conveying belt, according to a formula
Calculating the width of the damage; where Dis2 is the actual lesion width and len2 is the lesion width calculated from the preprocessed point cloud data.
7. The binocular line laser camera based conveyor belt longitudinal tear detection system of claim 1, further comprising: a back end control unit;
the input end of the rear-end control unit is connected with the output end of the data processing module, and the output end of the rear-end control unit is connected with the audible and visual alarm;
the back end control unit is configured to:
acquiring a feature extraction result corresponding to each frame of point cloud data and a feature extraction result of the whole target conveyor belt;
respectively comparing a feature extraction result corresponding to each frame of point cloud data and a feature extraction result of the whole target conveyor belt with alarm constraint conditions, and outputting an acousto-optic alarm instruction when the feature extraction result corresponding to the point cloud data and/or the feature extraction result of the whole target conveyor belt meet any one of the alarm constraint conditions; the alarm constraint conditions comprise three constraint conditions, namely a damage length threshold, a damage width threshold and a damage depth threshold.
8. The binocular line laser camera based conveyor belt longitudinal tear detection system of claim 7, further comprising: an audible and visual alarm;
and the audible and visual alarm is used for executing audible and visual alarm operation according to the received audible and visual alarm instruction.
9. The conveyor belt longitudinal tear detection system based on binocular line laser cameras of claim 1, wherein the binocular line laser cameras are installed at the bottom of the target conveyor belt in an obliquely upward 45 ° installation.
10. A method for detecting longitudinal tearing of a conveying belt based on a binocular line laser camera is characterized by comprising the following steps:
acquiring point cloud data of the bottom of a target conveyer belt; the point cloud data is acquired by a binocular line laser camera arranged at the bottom of the target conveyer belt;
carrying out damage characteristic information extraction on the point cloud data, and determining a characteristic extraction result corresponding to each frame of point cloud data; the feature extraction result comprises a non-damage feature result and a damage feature result; the lesion signature results include the location, category, and size of the lesion; the categories comprise tearing crack damage of the conveying belt and tearing deformation, distortion and folding damage of the conveying belt; the dimensions include one or more of length, width, depth;
and fusing the feature extraction results corresponding to the multi-frame point cloud data based on the transmission speed of the target conveyer belt so as to determine the feature extraction result of the whole target conveyer belt corresponding to the target conveyer belt running for one week.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210627532.2A CN114803386B (en) | 2022-06-06 | 2022-06-06 | Conveyor belt longitudinal tearing detection system and method based on binocular line laser camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210627532.2A CN114803386B (en) | 2022-06-06 | 2022-06-06 | Conveyor belt longitudinal tearing detection system and method based on binocular line laser camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114803386A true CN114803386A (en) | 2022-07-29 |
| CN114803386B CN114803386B (en) | 2023-08-25 |
Family
ID=82522174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210627532.2A Active CN114803386B (en) | 2022-06-06 | 2022-06-06 | Conveyor belt longitudinal tearing detection system and method based on binocular line laser camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114803386B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115690098A (en) * | 2022-12-16 | 2023-02-03 | 中科海拓(无锡)科技有限公司 | Method for detecting breakage and loss of iron wire |
| CN117208511A (en) * | 2023-11-08 | 2023-12-12 | 山西赛安自动控制有限公司 | Mining conveyer belt tearing detection device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102275723A (en) * | 2011-05-16 | 2011-12-14 | 天津工业大学 | Machine-vision-based online monitoring system and method for conveyer belt |
| CN104692078A (en) * | 2015-02-03 | 2015-06-10 | 赵磊 | Method for monitoring running state of belt-type conveyer |
| CN111661590A (en) * | 2020-06-08 | 2020-09-15 | 天地（常州）自动化股份有限公司 | Method for detecting tearing damage of conveying belt of mining belt conveyor |
| CN112213317A (en) * | 2020-09-28 | 2021-01-12 | 武汉科技大学 | Conveying belt tearing detection system based on three-dimensional laser scanning technology and detection method thereof |
| CN112325794A (en) * | 2020-10-12 | 2021-02-05 | 武汉万集信息技术有限公司 | Method, device and system for determining overall dimension of vehicle |
| CN112830189A (en) * | 2019-11-25 | 2021-05-25 | 山西戴德测控技术有限公司 | Conveyer belt tears prevention detection device based on binocular vision |
| WO2022101104A1 (en) * | 2020-11-16 | 2022-05-19 | Thyssenkrupp Industrial Solutions Ag | Method for identifying a distance-related running frictional resistance of a belt conveyor system |
| US20220171068A1 (en) * | 2020-12-02 | 2022-06-02 | Allstate Insurance Company | Damage detection and analysis using three-dimensional surface scans |
-
2022
- 2022-06-06 CN CN202210627532.2A patent/CN114803386B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102275723A (en) * | 2011-05-16 | 2011-12-14 | 天津工业大学 | Machine-vision-based online monitoring system and method for conveyer belt |
| CN104692078A (en) * | 2015-02-03 | 2015-06-10 | 赵磊 | Method for monitoring running state of belt-type conveyer |
| CN112830189A (en) * | 2019-11-25 | 2021-05-25 | 山西戴德测控技术有限公司 | Conveyer belt tears prevention detection device based on binocular vision |
| CN111661590A (en) * | 2020-06-08 | 2020-09-15 | 天地（常州）自动化股份有限公司 | Method for detecting tearing damage of conveying belt of mining belt conveyor |
| CN112213317A (en) * | 2020-09-28 | 2021-01-12 | 武汉科技大学 | Conveying belt tearing detection system based on three-dimensional laser scanning technology and detection method thereof |
| CN112325794A (en) * | 2020-10-12 | 2021-02-05 | 武汉万集信息技术有限公司 | Method, device and system for determining overall dimension of vehicle |
| WO2022101104A1 (en) * | 2020-11-16 | 2022-05-19 | Thyssenkrupp Industrial Solutions Ag | Method for identifying a distance-related running frictional resistance of a belt conveyor system |
| US20220171068A1 (en) * | 2020-12-02 | 2022-06-02 | Allstate Insurance Company | Damage detection and analysis using three-dimensional surface scans |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115690098A (en) * | 2022-12-16 | 2023-02-03 | 中科海拓(无锡)科技有限公司 | Method for detecting breakage and loss of iron wire |
| CN117208511A (en) * | 2023-11-08 | 2023-12-12 | 山西赛安自动控制有限公司 | Mining conveyer belt tearing detection device |
| CN117208511B (en) * | 2023-11-08 | 2024-02-02 | 山西赛安自动控制有限公司 | Mining conveyer belt tearing detection device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114803386B (en) | 2023-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114803386A (en) | Conveying belt longitudinal tearing detection system and method based on binocular line laser camera | |
| KR102022496B1 (en) | Process management and monitoring system using vision image detection and a method thereof | |
| CN111661590B (en) | Method for detecting tearing damage of conveying belt of mining belt conveyor | |
| AU2017349631B2 (en) | Belt inspection system and method | |
| CN114359246B (en) | Conveyor belt detection method, device, system, electronic device and medium | |
| CN105373135A (en) | Method and system for guiding airplane docking and identifying airplane type based on machine vision | |
| CN201890600U (en) | Machine vision belt tearing detecting device | |
| CN112528979B (en) | Transformer substation inspection robot obstacle distinguishing method and system | |
| CN114926395B (en) | Method and system for detecting infrared image drop strings of photovoltaic panel | |
| CN117557570B (en) | Rail vehicle abnormality detection method and system | |
| CN115144399A (en) | Assembly quality detection method and device based on machine vision | |
| CN104298993A (en) | Pole number positioning and identification method suitable for railway line in complex scene | |
| CN114445398A (en) | Method and device for monitoring state of side protection plate of hydraulic support of coal mining machine | |
| CN105447431A (en) | Docking airplane tracking and positioning method and system based on machine vision | |
| CN111626104B (en) | Cable hidden trouble point detection method and device based on unmanned aerial vehicle infrared thermal image | |
| CN113111728A (en) | Intelligent identification method and system for power production operation risk in transformer substation | |
| CN112288746A (en) | Machine vision-based off-square detection method and detection system | |
| CN110687548A (en) | Radar data processing system based on unmanned ship | |
| CN114426187B (en) | Data fusion and deep learning scraper chain early warning system | |
| CN116029987A (en) | Welding quality detection method and device and electronic equipment | |
| CN112446884A (en) | Method and device for positioning power transmission line in laser point cloud and terminal equipment | |
| CN113935946A (en) | Method and device for detecting underground obstacle in real time | |
| CN108181315B (en) | Image processing-based biscuit damage detection device and detection method | |
| CN113205133A (en) | Tunnel water stain intelligent identification method based on multitask learning | |
| CN111105395A (en) | A AI intelligence cloud platform for transmission of electricity operation monitoring |
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 |