CN115586749A - Workpiece machining track control method based on machine vision and related device - Google Patents
Workpiece machining track control method based on machine vision and related device Download PDFInfo
- Publication number
- CN115586749A CN115586749A CN202211587266.1A CN202211587266A CN115586749A CN 115586749 A CN115586749 A CN 115586749A CN 202211587266 A CN202211587266 A CN 202211587266A CN 115586749 A CN115586749 A CN 115586749A
- Authority
- CN
- China
- Prior art keywords
- processing
- track
- workpiece
- target
- deviation
- 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
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/4093—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by part programming, e.g. entry of geometrical information as taken from a technical drawing, combining this with machining and material information to obtain control information, named part programme, for the NC machine
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/36—Nc in input of data, input key till input tape
- G05B2219/36414—Compare image detected path with stored reference, difference corrects position
-
- 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
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Geometry (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Image Processing (AREA)
- Numerical Control (AREA)
Abstract
The invention relates to the field of artificial intelligence, and discloses a workpiece machining track control method based on machine vision and a related device, which are used for improving the accuracy of machining track control. The method comprises the following steps: generating a processing track of a target workpiece according to the target processing technology information and the workpiece information set to obtain a standard processing track; acquiring images of the actual machining state of the current workpiece to obtain a plurality of infrared images within machining detection time; carrying out initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified; performing track deviation calculation on a processing track to be recognized according to the standard processing track to obtain a corresponding track deviation data set; inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result; and matching the coping strategy to the track deviation detection result to obtain a target strategy, and transmitting the target strategy to the processing control terminal.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a workpiece machining track control method based on machine vision and a related device.
Background
The complexity of the curved surface of the target workpiece is higher and higher, and the requirements on the processing efficiency and stability of a control system are higher and higher; the machining requirement for the complex curved surface parts is higher in the fields of aviation, aerospace, energy, national defense and the like.
At present, the curvature speed limit is mostly calculated by using three adjacent points on a track to obtain an external circle, and the defects are that the numerical precision is sensitive when track points are dense, and the speed limit value during curvature reversing cannot be accurately processed, so that the accuracy rate is low during the analysis and control of the processing track of a workpiece.
Disclosure of Invention
The invention provides a workpiece machining track control method based on machine vision and a related device, which are used for improving the accuracy of machining track analysis control.
The invention provides a workpiece processing track control method based on machine vision, which comprises the following steps: acquiring a workpiece information set of a target workpiece from a preset database, and performing machining process matching on the target workpiece according to the workpiece information set to obtain target machining process information; generating a processing track of the target workpiece according to the target processing technology information and the workpiece information set to obtain a standard processing track; acquiring images of the current actual workpiece machining state based on preset machining detection time and a preset image acquisition terminal to obtain a plurality of infrared images within the machining detection time; performing initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified; performing track deviation calculation on the processing track to be identified according to the standard processing track to obtain a corresponding track deviation data set; inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result, wherein the track deviation detection result is track deviation and track non-deviation; and matching the corresponding strategies to the track deviation detection results to obtain corresponding target strategies, and transmitting the target strategies to a preset processing control terminal.
With reference to the first aspect, in a first implementation manner of the first aspect of the present invention, the generating a processing trajectory of the target workpiece according to the target processing technology information and the workpiece information set to obtain a standard processing trajectory includes: performing three-dimensional reconstruction through the structural information in the workpiece information set to obtain an initial workpiece three-dimensional model corresponding to the target workpiece; performing attribute assignment on the three-dimensional workpiece model through attribute information in the workpiece information set to obtain a target three-dimensional workpiece model; analyzing the processing area of the three-dimensional model of the target workpiece according to the target processing technology information to obtain a plurality of processing areas; and calibrating the processing tracks of the plurality of processing areas to generate standard processing tracks.
With reference to the first implementation manner of the first aspect, in a second implementation manner of the first aspect of the present invention, the performing a machining area analysis on the three-dimensional model of the target workpiece through the target machining process information to obtain a plurality of machining areas includes: matching the processing flows through the target processing process information to obtain the corresponding workpiece processing flow; and dividing the processing area of the three-dimensional model of the target workpiece based on the workpiece processing flow to obtain a plurality of processing areas.
With reference to the first aspect, in a third implementation manner of the first aspect of the present invention, the performing an initial processing trajectory analysis on the multiple infrared images to obtain a processing trajectory to be identified includes: carrying out binarization processing on the plurality of infrared images to obtain a plurality of corresponding binarization images; performing heat point analysis on the plurality of binary images to obtain corresponding heat point sets; based on a preset space coordinate system, carrying out coordinate conversion on each heat point in the heat point set to obtain a heat point coordinate set; and carrying out normalization processing on the heat point coordinate set to obtain a processing track to be identified.
With reference to the first aspect, in a fourth implementation manner of the first aspect of the present invention, the performing a trajectory deviation calculation on the to-be-identified processing trajectory according to the standard processing trajectory to obtain a corresponding trajectory deviation data set includes: performing curve mapping processing on the standard processing track to obtain a corresponding first processing track curve; carrying out curve mapping processing on the processing track to be identified to obtain a corresponding second processing track curve; performing displacement deviation analysis on the first processing track curve and the second processing track curve to obtain corresponding track deviation data sets;
with reference to the fourth implementation manner of the first aspect, in a fifth implementation manner of the first aspect of the present invention, the performing displacement deviation analysis on the first processing trajectory curve and the second processing trajectory curve to obtain a corresponding trajectory deviation data set includes: performing curve difference calculation on the first processing track curve and the second processing track curve to obtain a corresponding difference calculation result set; respectively screening each difference calculation result in the difference calculation result set based on a preset deviation displacement threshold value to obtain a first data set exceeding the deviation displacement threshold value and a second data set not exceeding the deviation displacement threshold value; and performing ratio calculation on the first data set and the second data set to obtain a ratio result, and determining a corresponding track offset data set according to the ratio result.
With reference to the first aspect, in a sixth implementation manner of the first aspect of the present invention, the performing policy matching on the track deviation detection result to obtain a corresponding target policy, and transmitting the target policy to a preset processing control terminal includes: on the basis of a preset coping strategy database, coping strategies are collected through the workpiece information set to obtain a coping strategy set; performing word segmentation processing on the track deviation detection result to obtain a corresponding word segmentation result; and matching keywords according to the word segmentation result and the coping strategy set to obtain a corresponding target strategy and transmitting the target strategy to a preset processing control terminal.
The second aspect of the present invention provides a workpiece processing track control method and device based on machine vision, where the workpiece processing track control method and device based on machine vision includes:
the acquisition module is used for acquiring a workpiece information set of a target workpiece from a preset database, and performing machining process matching on the target workpiece according to the workpiece information set to obtain target machining process information;
the generating module is used for generating a processing track of the target workpiece according to the target processing technology information and the workpiece information set to obtain a standard processing track;
the acquisition module is used for acquiring images of the current workpiece processing actual state based on preset processing detection time and a preset image acquisition terminal to obtain a plurality of infrared images within the processing detection time;
the analysis module is used for carrying out initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified;
the calculation module is used for carrying out track deviation calculation on the processing track to be identified according to the standard processing track to obtain a corresponding track deviation data set;
the detection module is used for inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result, wherein the track deviation detection result is track deviation and track non-deviation;
and the matching module is used for matching the corresponding strategy of the track deviation detection result to obtain a corresponding target strategy and transmitting the target strategy to a preset processing control terminal.
The invention provides a workpiece processing track control method and device based on machine vision, which comprises the following steps: a memory and at least one processor, the memory having instructions stored therein; the at least one processor calls the instructions in the memory to cause the machine vision-based workpiece processing track control method device to execute the machine vision-based workpiece processing track control method.
A fourth aspect of the present invention provides a computer-readable storage medium having stored therein instructions, which, when run on a computer, cause the computer to execute the above-mentioned machine vision-based workpiece processing trajectory control method.
According to the technical scheme provided by the invention, the processing track of the target workpiece is generated according to the target processing technology information and the workpiece information set, and a standard processing track is obtained; acquiring images of the actual machining state of the current workpiece to obtain a plurality of infrared images within machining detection time; carrying out initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified; performing track deviation calculation on a to-be-identified processing track according to the standard processing track to obtain a corresponding track deviation data set; inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result; the processing method and the processing system have the advantages that the processing track is detected in real time in the processing process of the workpiece, and the processing process is dynamically adjusted, so that the track deviation detection accuracy is improved, and further the accuracy of processing track control is improved.
Drawings
Fig. 1 is a schematic diagram of an embodiment of a workpiece processing trajectory control method based on machine vision according to an embodiment of the present invention;
FIG. 2 is a flowchart of an exemplary process for analyzing an initial processing trajectory of a plurality of infrared images;
FIG. 3 is a flowchart of a process of analyzing displacement deviation of a processing trajectory to be recognized according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating a process of performing a displacement deviation analysis on a first processing trajectory curve and a second processing trajectory curve according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an embodiment of a method and an apparatus for controlling a processing trajectory of a workpiece based on machine vision according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an embodiment of a workpiece processing trajectory control method and apparatus based on machine vision in an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a workpiece machining track control method, a workpiece machining track control device, workpiece machining track control equipment and a storage medium based on machine vision, which are used for improving the accuracy of machining track analysis control. The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments described herein are capable of operation in sequences other than those illustrated or otherwise described herein. Moreover, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a detailed flow of an embodiment of the present invention is described below, and referring to fig. 1, an embodiment of a method for controlling a processing trajectory of a workpiece based on machine vision in an embodiment of the present invention includes:
s101, acquiring a workpiece information set of a target workpiece from a preset database, and performing machining process matching on the target workpiece according to the workpiece information set to obtain target machining process information;
it is to be understood that the executing subject of the present invention may be a workpiece processing trajectory control method device based on machine vision, and may also be a terminal or a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.
Specifically, the database is used for realizing the construction and storage of the processing technology information of the workpiece, wherein the construction of the processing technology information comprises characteristic identification and the establishment of a workpiece information set, and an intelligent algorithm is adopted to identify the technology sequence and generate the processing technology information; and the processing technology information based on the workpiece information characteristic recognition result realizes the calculation of the workpiece information characteristic recognition result of the workpiece technology characteristic and the selection of the processing technology information, thereby constructing and obtaining a workpiece information set. The embodiment carries out characteristic analysis on a target workpiece, including precision characteristic identification, size identification and shape identification; sorting and classifying the identified characteristic information to generate a complete workpiece information set; selecting one or more proper processing process information from a process information base; and further improving the selected process information, generating complete processing process information according to the characteristic information of the target workpiece, allowing different processing process information of the plurality of process information to exist, and matching the processing process of the target workpiece according to the workpiece information set to obtain the target processing process information. Furthermore, different constraints are added according to various conditions in the target processing technology information making process to ensure that the automatically generated processing technology information has feasibility; automatically rearranging the process sequence of each candidate processing process information by adopting an intelligent algorithm according to the constraint condition; and generating complete target machining process information by the different process sequences of each candidate machining process information.
S102, generating a processing track of a target workpiece according to the target processing technology information and the workpiece information set to obtain a standard processing track;
specifically, firstly, target processing technology information of a target workpiece is obtained, and production technology information of the target workpiece is determined according to the target processing technology information; secondly, configuring processing structure information corresponding to the target workpiece according to the production process information, and acquiring processing track information corresponding to the processing structure information; then establishing an initial workpiece three-dimensional model of the target workpiece based on the production process information and the processing track information; and finally, carrying out model attribute assignment based on the initial workpiece three-dimensional model to obtain a target workpiece three-dimensional model, and then processing the target workpiece according to the target workpiece three-dimensional model. And determining the production process information of the target workpiece based on the target processing process information, configuring corresponding processing structure information according to the production process information, further acquiring the processing track information of the processing structure information, performing attribute assignment, and constructing a three-dimensional model of the target workpiece. The method comprises the steps of analyzing the machining areas according to the established three-dimensional model of the target workpiece to obtain a plurality of machining areas, calibrating the machining track to generate a standard machining track.
S103, acquiring images of the current actual workpiece machining state based on preset machining detection time and a preset image acquisition terminal to obtain a plurality of infrared images within the machining detection time;
specifically, the current workpiece processing actual state is subjected to image acquisition based on preset processing detection time and a preset image acquisition terminal, and acquired original images are preprocessed, wherein the preprocessing comprises image histogram equalization and denoising. The image histogram equalization includes: firstly, performing histogram analysis on an image along a first preset direction, if an area lower than a set gray threshold exists, indicating that a part of a target workpiece is found, entering an outline calculation step, if the area lower than the set threshold is not found, performing bidirectional search, and if the area lower than the set threshold is not found in the bidirectional search, prompting that the target workpiece is not found; searching in the first preset direction according to a search strategy in the second preset direction; and obtaining the balanced image through the two searches. The image denoising method comprises the following steps: firstly, gaussian filtering is carried out on an image to obtain a noise-reduced image, then the image is compared with the image subjected to Gaussian filtering to obtain a model, and the model is used as an image extracted by a Canny operator contour; after the contour is obtained, calculating the area of the contour, then theoretically calculating the area of the magnetic shoe, wherein the theoretical area is calculated only once and then stored in a database for next calling; if this parameter already exists in the database, it is called directly. By preprocessing the acquired original image in two steps of image histogram equalization and denoising, the problem of poor track analysis stability caused by the fact that a target workpiece detection result is easily influenced by image noise points and uneven image gray level distribution is solved, and therefore stability and accuracy of track analysis are greatly improved.
S104, performing initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified;
specifically, according to a plurality of infrared images input by the trained trajectory analysis model, binarization processing is performed on the plurality of infrared images to obtain an image after binarization processing. Since the resolution of the input infrared images cannot be determined, and the input dimension of the trajectory analysis model is usually fixed, in order to enable the trajectory analysis model to normally process the infrared images, binarization processing is performed on the infrared images according to the resolutions of the infrared images and the input dimension of the trajectory analysis model. The track analysis model comprises three layers of convolution networks, a pooling layer and two layers of convolution networks. And inputting the image after the binarization processing into a track analysis model, and performing convolution processing on the convolution layers of the preset number of the track analysis model to obtain the image characteristics after the convolution processing of the convolution layers of the preset number. The method comprises the steps of carrying out heat point analysis on a plurality of binary images to obtain corresponding heat point sets, carrying out coordinate conversion on each heat point in the heat point sets based on a preset space coordinate system to obtain heat point coordinate sets, determining position areas of a target object in a plurality of infrared images based on the heat point coordinate sets, carrying out coordinate conversion on each heat point in the heat point sets based on the position areas to obtain heat point coordinate sets, and carrying out normalization processing on the heat point coordinate sets to obtain a processing track to be recognized.
S105, performing track deviation calculation on the processing track to be recognized according to the standard processing track to obtain a corresponding track deviation data set;
specifically, a preset standard processing track is inquired from a database according to a workpiece information set of a target workpiece, track deviation calculation is carried out on the processing track to be identified according to the standard processing track, the target workpiece is scanned to obtain three-dimensional point cloud data, the processing track to be identified is extracted according to an algorithm and subjected to coordinate transformation, the processing track to be identified is converted to a robot end effector coordinate system, and then the target workpiece and the processing track to be identified are coded and stored in the database; acquiring a processing track to be identified: acquiring experimental data in a processing scene and processing data in a track extraction scene through network communication, then compiling a calibration algorithm to calculate a coordinate system transformation matrix, then carrying out track offset calculation on a standard processing track and a processing track to be identified to obtain a plurality of track offset point data, and carrying out set conversion on the plurality of track offset point data to obtain a track offset data set. For a target workpiece, the automatic processing production line can process the same type of workpiece only by acquiring a processing track once, and if the target workpiece on the automatic processing production line of the robot is replaced, the target workpiece is scanned in a track extraction scene to extract the processing track and perform coordinate conversion, and then the processing track is coded and stored; and the upper computer in the robot processing scene accesses the database through network communication, selects corresponding processing track data and transmits the corresponding processing track data to a specified robot processing production line to complete a processing task.
S106, inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result, wherein the track deviation detection result is track deviation and track non-deviation;
specifically, inputting the trajectory offset data set into a generation network in the trajectory offset detection model, wherein the generation network includes: a double-layer GRU network, a two-layer full-connection network and an output layer; performing feature coding on the track offset data set through a double-layer GRU network to obtain target coding features; performing characteristic operation on the target coding characteristics through a two-layer fully-connected network to obtain characteristic operation information; and inputting the characteristic operation information into an output layer for conversion to obtain a track deviation detection result, wherein the track deviation detection result is track deviation and track non-deviation. In this embodiment, the dual-layer GRU network is composed of two layers of GRUs, the first layer is a command layer formed by connecting one unidirectional GRU, 256 GRU units are provided, the feature vector output by the first layer is input into each GRU unit connected with the first group of 16 unidirectional GRUs in the corresponding divergence layer, the first GRU unit in the divergence layer outputs an integer value, the second GRU unit outputs an integer value, \8230 \ 8230, the 16 GRU unit outputs an integer value, all the output integer values are combined into a value vector, wherein the output layer outputs a probability value, the range of the probability value is [0,1], when the probability value is [0,0.5 ], the track deviation detection result is determined to be a track deviation, and when the probability value is [0.5,1], the track deviation detection result is determined to be no track deviation.
And S107, performing corresponding strategy matching on the track deviation detection result to obtain a corresponding target strategy, and transmitting the target strategy to a preset processing control terminal.
Specifically, the element information of the first target policy to be processed is obtained according to the track deviation detection result. And entering a second target strategy preset in a coping strategy database traversal library, judging whether traversal of all target strategies in the coping strategy database is finished or not, and taking out the target strategies from the coping strategy database if the traversal of all target strategies in the coping strategy database is not finished. Whether the target object of the first target policy overlaps with the target object (specifically, the target object) of the target policy in the policy database is judged, and if the target object of the first target policy overlaps with the target object (specifically, the target object) of the target policy in the policy database, the first target policy is added into the candidate set as the target policy corresponding to the trajectory offset detection result in the embodiment, and the element information of the target policy, that is, the target object, is marked. Judging whether the conversion situation of the first target strategy is overlapped with the conversion situation of the target strategy in the coping strategy database, if so, adding the target strategy into the candidate set and marking the conversion situation of the target strategy, sorting the target strategies in the candidate set according to the number of marked elements and then outputting, returning the corresponding target strategies, and transmitting the target strategies to a preset processing control terminal.
In the embodiment of the invention, the processing track of the target workpiece is generated according to the target processing technology information and the workpiece information set to obtain a standard processing track; acquiring images of the actual machining state of the current workpiece to obtain a plurality of infrared images within machining detection time; carrying out initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified; performing track deviation calculation on a processing track to be recognized according to the standard processing track to obtain a corresponding track deviation data set; inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result; the track deviation detection result is matched with the coping strategy to obtain the target strategy, and the target strategy is transmitted to the processing control terminal.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Performing three-dimensional reconstruction through structural information in the workpiece information set to obtain an initial workpiece three-dimensional model corresponding to the target workpiece;
(2) Performing attribute assignment on the three-dimensional workpiece model through attribute information in the workpiece information set to obtain a target three-dimensional workpiece model;
(3) Analyzing the processing area of the three-dimensional model of the target workpiece through the target processing process information to obtain a plurality of processing areas;
(4) And calibrating the processing tracks of the plurality of processing areas to generate standard processing tracks.
Specifically, the server inputs structural point cloud information of structural information in the workpiece information set into a preset first training model, wherein the first training model comprises: the system comprises a double-layer point cloud convolution network, a batch normalization layer, a feature fusion layer and a classification network; the server extracts the characteristics of the structural point cloud information through a double-layer point cloud convolution network to obtain first characteristic information; the server inputs the first characteristic information into a batch normalization layer for normalization processing to obtain normalized characteristic information, the first characteristic information generated from each convolution layer in the second layer point cloud convolution network is subjected to batch normalization processing through the batch normalization layer, and the data of the characteristic information is subjected to normalized processing to improve the convergence of the model; the server inputs the normalized feature information into a feature fusion layer to perform feature fusion processing to obtain second feature information, the feature fusion layer is used for refining the fine granularity of the target feature information and expanding the detection precision, the last individual pooling layer of a second layer point cloud convolution network is connected with the last convolution layer, the feature information after corresponding pooling or convolution is overlapped, the resolution of the features is further expanded, and the fine granularity of the features is refined; the server inputs the second feature information into a classification network, predicts a plurality of pixel points in the second feature information through the classification network, and generates a prediction result corresponding to the second feature information, wherein the classification layer is a plurality of 1x1 convolution layers and is used for predicting the position of a feature frame in the second feature information, the output data format of the classification layer is a priori frame number x (5 + classification number), in a first training model, the priori frame number is 5, the classification number is 20 characters, and 5 in brackets represents a central two-dimensional coordinate, a width, a height and a confidence coefficient of a feature information boundary frame, wherein the confidence coefficient of the feature information boundary frame is represented by an IOU (Intersection Unit, intersection ratio); and the server optimizes the parameters of the first training model according to the prediction result and the structural point cloud information until the first training model is converged to obtain a three-dimensional reconstruction model. Performing attribute assignment on the three-dimensional workpiece model through attribute information in the workpiece information set to obtain a target three-dimensional workpiece model; processing area analysis is carried out on the target workpiece three-dimensional model through the target processing technology information to obtain a plurality of processing areas; and respectively extracting track points in the plurality of processing areas, then calibrating the processing tracks of the plurality of processing areas, and synthesizing the calibrated processing tracks to generate a standard processing track.
Further, the server inputs the structural point cloud information into a first layer point cloud convolution network in a double-layer point cloud convolution network, dimension clustering is carried out on labels in the structural point cloud information through the first layer point cloud convolution network, a priori frame corresponding to the structural point cloud information is generated, the first layer point cloud convolution network is a classification network, and the area of the prediction frame is determined through dimension clustering on the label information, so that the priori frame is obtained. The priori frame is used for defining the central point and the size of a target characteristic region, so that the learning of the second layer point cloud convolution network is easier; the server inputs the structural point cloud information with the prior frame into a second layer point cloud convolution network in the double-layer point cloud convolution network, extracts first characteristic information of the structural point cloud information through the second layer point cloud convolution network, and adjusts the position range of the prior frame, wherein the second layer point cloud convolution network is obtained by modifying a classification model of the first layer point cloud convolution network, and the difference is that the former is a detection model training network, and the latter is a classification model training network. Inputting the structural point cloud information into the network, and extracting corresponding characteristic information; and then, after each convolution, the characteristics of the characteristic information are deepened, refined and corrected.
In a specific embodiment, the process of analyzing the processing area of the three-dimensional model of the target workpiece according to the target processing process information to obtain a plurality of processing areas may specifically include the following steps:
(1) Matching the processing flows through the target processing process information to obtain the corresponding workpiece processing flow;
(2) And dividing the processing area of the three-dimensional model of the target workpiece based on the workpiece processing flow to obtain a plurality of processing areas.
Specifically, the processing flow matching is carried out through the target processing technology information to obtain a corresponding workpiece processing flow, the processing area division is carried out on the target workpiece three-dimensional model based on the workpiece processing flow to obtain a plurality of processing areas, and a plurality of area information including the target workpiece three-dimensional model is obtained; adding a plurality of area identifications on the surface of the three-dimensional model of the target workpiece; according to the area identifications, carrying out coincidence matching on the surface of the target workpiece three-dimensional model and a preset coordinate system; and extracting and obtaining a plurality of processing areas according to the surface of the superposed and matched three-dimensional model of the target workpiece.
In a specific embodiment, as shown in fig. 2, the process of executing step S105 may specifically include the following steps:
s201, carrying out binarization processing on the plurality of infrared images to obtain a plurality of corresponding binarization images;
s202, carrying out heat point analysis on the plurality of binary images to obtain corresponding heat point sets;
s203, based on a preset space coordinate system, carrying out coordinate conversion on each heat point in the heat point set to obtain a heat point coordinate set;
and S204, normalizing the heat point coordinate set to obtain the processing track to be identified.
Specifically, the server performs binarization processing on a plurality of infrared images to obtain a plurality of corresponding binarization images; performing heat point analysis on the multiple binary images to obtain corresponding heat point sets, wherein a plurality of coordinate points with probability values larger than a preset threshold value are determined on each heat map of the multiple binary images; determining pixel points of the plurality of coordinate points in the plurality of infrared images respectively; determining a candidate frame set corresponding to the target object in the plurality of infrared images based on pixel points corresponding to the plurality of coordinate points in the plurality of infrared images respectively to obtain a candidate frame set corresponding to at least one heat map respectively; and determining a corresponding heat point set of the target workpiece in the plurality of infrared images based on the candidate frame set corresponding to each of the at least one heat map. Combining the candidate frame sets corresponding to the at least one heat map based on a non-maximum suppression algorithm to obtain a candidate frame corresponding to the at least one heat map; and combining the candidate frames corresponding to at least one heat map based on a non-maximum suppression algorithm, determining the combined candidate frames as a heat point set of the target object in a plurality of infrared images, performing coordinate conversion on each heat point in the heat point set to obtain a heat point coordinate set, and performing normalization processing on the heat point coordinate set to obtain the processing track to be identified.
In a specific embodiment, as shown in fig. 3, the process of executing step S106 may specifically include the following steps:
s301, performing curve mapping processing on the standard processing track to obtain a corresponding first processing track curve;
s302, carrying out curve mapping processing on the processing track to be identified to obtain a corresponding second processing track curve;
and S303, carrying out displacement deviation analysis on the first processing track curve and the second processing track curve to obtain corresponding track deviation data sets.
Specifically, curve mapping processing is carried out on the standard processing track to obtain a corresponding first processing track curve, and a curve fitting function is utilized to obtain a track curve corresponding to the standard processing track and the processing track to be identified; carrying out curve mapping processing on the processing track to be identified to obtain a corresponding second processing track curve; carrying out displacement deviation analysis on the first processing track curve and the second processing track curve to obtain a corresponding track deviation data set, specifically: extracting a plurality of curve characteristic values of the first processing track curve and the second processing track curve, then carrying out characteristic point matching on the plurality of curve characteristic values to obtain a plurality of curve characteristic value pairs, carrying out difference calculation on the plurality of curve characteristic value pairs to obtain a corresponding difference calculation result set, and finally comparing according to the corresponding difference calculation result set to generate a track deviation data set.
In a specific embodiment, as shown in fig. 4, the process of executing step S303 may specifically include the following steps:
s401, carrying out curve difference calculation on the first processing track curve and the second processing track curve to obtain a corresponding difference calculation result set;
s402, screening each difference calculation result in the difference calculation result set based on a preset deviation displacement threshold value to obtain a first data set exceeding the deviation displacement threshold value and a second data set not exceeding the deviation displacement threshold value;
s403, ratio calculation is carried out on the first data set and the second data set to obtain a ratio result, and a corresponding track deviation data set is determined according to the ratio result.
Specifically, curve difference calculation is carried out on the first processing track curve and the second processing track curve to obtain a corresponding difference calculation result set; respectively screening each difference calculation result in the difference calculation result set based on a preset deviation displacement threshold value to obtain a first data set exceeding the deviation displacement threshold value and a second data set not exceeding the deviation displacement threshold value, wherein the deviation displacement threshold value is an average value corresponding to all the difference calculation results in the difference calculation result set; and carrying out ratio calculation on the first data set and the second data set to obtain a ratio result, and determining a corresponding track offset data set according to the ratio result, wherein the track offset data set comprises a plurality of data points, classifying the data points according to a plurality of processing areas, and then carrying out set combination on the classified data points to generate a track offset data set.
In a specific embodiment, the step S107 is executed, and may include the following steps:
(1) On the basis of a preset coping strategy database, coping strategies are collected through a workpiece information set to obtain a coping strategy set;
(2) Performing word segmentation processing on the track deviation detection result to obtain a corresponding word segmentation result;
(3) And matching keywords according to the word segmentation result and the coping strategy set to obtain a corresponding target strategy and transmitting the target strategy to a preset processing control terminal.
Specifically, word segmentation processing is performed on the track deviation detection result to obtain a corresponding word segmentation result, specifically, each word segmentation in an initial word segmentation set is adopted to obtain a weight of the word segmentation, a distance from the word segmentation to a word segmentation closest to the target word segmentation set is determined, the weight of the word segmentation is obtained for each word segmentation in the target word segmentation set, the distance from the word segmentation to the word segmentation closest to the initial word segmentation set is determined, a word segmentation result is obtained according to the weight and the corresponding distance corresponding to each word segmentation in the initial word segmentation set and the weight and the corresponding distance corresponding to each word segmentation in the target word segmentation set, keyword matching is performed according to the word segmentation result and a coping strategy set to obtain a corresponding target strategy, and the target strategy is transmitted to a preset processing control terminal.
Referring to fig. 5, the method for controlling a workpiece machining trajectory based on machine vision in the embodiment of the present invention is described above, and a device for controlling a workpiece machining trajectory based on machine vision in the embodiment of the present invention is described below, in which one embodiment of the device for controlling a workpiece machining trajectory based on machine vision in the embodiment of the present invention includes:
an obtaining module 501, configured to obtain a workpiece information set of a target workpiece from a preset database, and perform processing technology matching on the target workpiece according to the workpiece information set to obtain target processing technology information;
a generating module 502, configured to generate a processing trajectory of the target workpiece according to the target processing technology information and the workpiece information set, so as to obtain a standard processing trajectory;
the acquisition module 503 is configured to perform image acquisition on the current actual processing state of the workpiece based on preset processing detection time and a preset image acquisition terminal, so as to obtain a plurality of infrared images within the processing detection time;
an analysis module 504, configured to perform initial processing trajectory analysis on the multiple infrared images to obtain a processing trajectory to be identified;
a calculating module 505, configured to perform trajectory deviation calculation on the processing trajectory to be identified according to the standard processing trajectory to obtain a corresponding trajectory deviation data set;
a detection module 506, configured to input the trajectory offset data set into a preset trajectory offset detection model to perform trajectory offset detection, so as to obtain a trajectory offset detection result, where the trajectory offset detection result is a trajectory offset and a trajectory non-offset;
and the matching module 507 is configured to perform policy matching on the track deviation detection result to obtain a corresponding target policy, and transmit the target policy to a preset processing control terminal.
Generating a processing track of a target workpiece according to the target processing technology information and the workpiece information set through the cooperative cooperation of all the components to obtain a standard processing track; acquiring images of the actual machining state of the current workpiece to obtain a plurality of infrared images within machining detection time; carrying out initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified; performing track deviation calculation on a to-be-identified processing track according to the standard processing track to obtain a corresponding track deviation data set; inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result; the processing method and the processing system have the advantages that the processing track is detected in real time in the processing process of the workpiece, and the processing process is dynamically adjusted, so that the track deviation detection accuracy is improved, and further the accuracy of processing track control is improved.
Fig. 6 is a schematic structural diagram of a workpiece processing trajectory control method apparatus based on machine vision according to an embodiment of the present invention, the workpiece processing trajectory control method apparatus based on machine vision 600 may generate relatively large differences due to different configurations or performances, and may include one or more processors (CPUs) 610 (e.g., one or more processors) and a memory 620, and one or more storage media 630 (e.g., one or more mass storage devices) storing applications 633 or data 632. Memory 620 and storage medium 630 may be, among other things, transitory or persistent storage. The program stored on the storage medium 630 may include one or more modules (not shown), each of which may include a series of instruction operations in the machine vision-based workpiece processing trajectory control method apparatus 600. Still further, the processor 610 may be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the machine vision based workpiece processing trajectory control method apparatus 600.
The machine vision-based workpiece processing trajectory control method apparatus 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input-output interfaces 660, and/or one or more operating systems 631, such as Windows Server, mac OS X, unix, coping policy database inux, freeBSD, and so forth. Those skilled in the art will appreciate that the machine vision based workpiece processing trajectory control method apparatus configuration shown in fig. 6 does not constitute a limitation of the machine vision based workpiece processing trajectory control method apparatus and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
The invention also provides a workpiece machining track control method and device based on machine vision, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, enable the processor to execute the steps of the workpiece machining track control method based on machine vision in the above embodiments.
The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium, and which may also be a volatile computer-readable storage medium, having stored therein instructions, which, when executed on a computer, cause the computer to execute the steps of the machine vision-based workpiece processing trajectory control method.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present invention, which are essential or part of the technical solutions contributing to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A workpiece processing track control method based on machine vision is characterized by comprising the following steps:
acquiring a workpiece information set of a target workpiece from a preset database, and performing machining process matching on the target workpiece according to the workpiece information set to obtain target machining process information;
generating a processing track of the target workpiece according to the target processing technology information and the workpiece information set to obtain a standard processing track;
acquiring images of the current actual processing state of the workpiece based on preset processing detection time and a preset image acquisition terminal to obtain a plurality of infrared images within the processing detection time;
performing initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified;
performing track deviation calculation on the processing track to be identified according to the standard processing track to obtain a corresponding track deviation data set;
inputting the track deviation data set into a preset track deviation detection model for track deviation detection to obtain a track deviation detection result, wherein the track deviation detection result is track deviation and track non-deviation;
and matching the corresponding strategies to the track deviation detection results to obtain corresponding target strategies, and transmitting the target strategies to a preset processing control terminal.
2. The method for controlling the processing track of the workpiece based on the machine vision as claimed in claim 1, wherein the generating the processing track of the target workpiece according to the target processing technology information and the set of workpiece information to obtain a standard processing track comprises:
performing three-dimensional reconstruction through the structural information in the workpiece information set to obtain an initial workpiece three-dimensional model corresponding to the target workpiece;
performing attribute assignment on the workpiece three-dimensional model through attribute information in the workpiece information set to obtain a target workpiece three-dimensional model;
processing area analysis is carried out on the target workpiece three-dimensional model through the target processing technology information to obtain a plurality of processing areas;
and calibrating the processing tracks of the plurality of processing areas to generate standard processing tracks.
3. The method for controlling the processing track of the workpiece based on the machine vision as claimed in claim 2, wherein the step of analyzing the three-dimensional model of the target workpiece by the target processing technology information to obtain a plurality of processing areas comprises:
matching the processing flows according to the target processing process information to obtain the corresponding workpiece processing flow;
and dividing the processing area of the target workpiece three-dimensional model based on the workpiece processing flow to obtain a plurality of processing areas.
4. The workpiece processing track control method based on machine vision according to claim 1, wherein the performing initial processing track analysis on the plurality of infrared images to obtain the processing track to be identified comprises:
carrying out binarization processing on the plurality of infrared images to obtain a plurality of corresponding binarization images;
performing heat point analysis on the plurality of binary images to obtain corresponding heat point sets;
based on a preset space coordinate system, carrying out coordinate conversion on each heat point in the heat point set to obtain a heat point coordinate set;
and carrying out normalization processing on the heat point coordinate set to obtain a processing track to be identified.
5. The machine-vision-based workpiece processing track control method according to claim 1, wherein the performing track deviation calculation on the processing track to be identified according to the standard processing track to obtain a corresponding track deviation data set comprises:
performing curve mapping processing on the standard processing track to obtain a corresponding first processing track curve;
carrying out curve mapping processing on the processing track to be identified to obtain a corresponding second processing track curve;
and carrying out displacement deviation analysis on the first processing track curve and the second processing track curve to obtain a corresponding track deviation data set.
6. The method of claim 5, wherein performing a displacement deviation analysis on the first machining trajectory curve and the second machining trajectory curve to obtain a corresponding trajectory deviation data set comprises:
performing curve difference calculation on the first processing track curve and the second processing track curve to obtain a corresponding difference calculation result set;
respectively screening each difference calculation result in the difference calculation result set based on a preset deviation displacement threshold value to obtain a first data set exceeding the deviation displacement threshold value and a second data set not exceeding the deviation displacement threshold value;
and performing ratio calculation on the first data set and the second data set to obtain a ratio result, and determining a corresponding track offset data set according to the ratio result.
7. The machine-vision-based workpiece processing trajectory control method of claim 1, wherein the performing of the corresponding policy matching on the trajectory deviation detection result to obtain a corresponding target policy, and transmitting the target policy to a preset processing control terminal comprises:
on the basis of a preset coping strategy database, coping strategies are collected through the workpiece information set to obtain a coping strategy set;
performing word segmentation processing on the track deviation detection result to obtain a corresponding word segmentation result;
and matching keywords according to the word segmentation result and the coping strategy set to obtain a corresponding target strategy and transmitting the target strategy to a preset processing control terminal.
8. A workpiece processing track control method device based on machine vision is characterized in that the workpiece processing track control method device based on machine vision comprises the following steps:
the acquisition module is used for acquiring a workpiece information set of a target workpiece from a preset database, and performing machining process matching on the target workpiece according to the workpiece information set to obtain target machining process information;
the generating module is used for generating a processing track of the target workpiece according to the target processing technology information and the workpiece information set to obtain a standard processing track;
the acquisition module is used for acquiring images of the current actual machining state of the workpiece based on preset machining detection time and a preset image acquisition terminal to obtain a plurality of infrared images within the machining detection time;
the analysis module is used for carrying out initial processing track analysis on the plurality of infrared images to obtain a processing track to be identified;
the calculation module is used for carrying out track deviation calculation on the processing track to be identified according to the standard processing track to obtain a corresponding track deviation data set;
the detection module is used for inputting the track offset data set into a preset track offset detection model for track offset detection to obtain a track offset detection result, wherein the track offset detection result is track deviation and track non-deviation;
and the matching module is used for matching the corresponding strategy of the track deviation detection result to obtain a corresponding target strategy and transmitting the target strategy to a preset processing control terminal.
9. The device for controlling the machining track of the workpiece based on the machine vision is characterized by comprising the following steps: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invokes the instructions in the memory to cause the machine vision based workpiece processing trajectory control method apparatus to perform the machine vision based workpiece processing trajectory control method of any of claims 1-7.
10. A computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the machine-vision based workpiece processing trajectory control method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211587266.1A CN115586749B (en) | 2022-12-12 | 2022-12-12 | Workpiece machining track control method based on machine vision and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211587266.1A CN115586749B (en) | 2022-12-12 | 2022-12-12 | Workpiece machining track control method based on machine vision and related device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115586749A true CN115586749A (en) | 2023-01-10 |
| CN115586749B CN115586749B (en) | 2023-03-21 |
Family
ID=84783144
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211587266.1A Active CN115586749B (en) | 2022-12-12 | 2022-12-12 | Workpiece machining track control method based on machine vision and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115586749B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115971378A (en) * | 2023-03-16 | 2023-04-18 | 天津丰通晟源科技有限公司 | Special-shaped spring production control method and system based on visual detection |
| CN116518868A (en) * | 2023-07-05 | 2023-08-01 | 深圳市海塞姆科技有限公司 | Deformation measurement method, device, equipment and storage medium based on artificial intelligence |
| CN117146709A (en) * | 2023-10-30 | 2023-12-01 | 钛玛科（北京）工业科技有限公司 | Deviation rectifying control device and system based on automatic selection of sensor |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102486641A (en) * | 2010-12-03 | 2012-06-06 | 中国科学院沈阳自动化研究所 | Artificial-teeth processing route generating method |
| CN103336485A (en) * | 2013-06-18 | 2013-10-02 | 南京航空航天大学 | Rapid generating method of milling path of web of airplane structural member |
| CN103365243A (en) * | 2013-06-18 | 2013-10-23 | 南京航空航天大学 | Method for rapidly generating corner side milling process path |
| CN111014892A (en) * | 2019-12-13 | 2020-04-17 | 华中科技大学鄂州工业技术研究院 | Welding seam track monitoring system |
| CN111097664A (en) * | 2019-12-20 | 2020-05-05 | 广西柳州联耕科技有限公司 | Real-time deviation rectifying method based on robot gluing |
| CN111192307A (en) * | 2019-12-20 | 2020-05-22 | 广西柳州联耕科技有限公司 | Self-adaptive deviation rectifying method based on laser cutting of three-dimensional part |
| CN113172307A (en) * | 2021-03-24 | 2021-07-27 | 苏州奥天智能科技有限公司 | Industrial robot system of visual module based on laser and visible light fusion |
| CN113634964A (en) * | 2021-08-25 | 2021-11-12 | 武汉理工大学 | Gantry type robot welding equipment and welding process for large-sized component |
-
2022
- 2022-12-12 CN CN202211587266.1A patent/CN115586749B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102486641A (en) * | 2010-12-03 | 2012-06-06 | 中国科学院沈阳自动化研究所 | Artificial-teeth processing route generating method |
| CN103336485A (en) * | 2013-06-18 | 2013-10-02 | 南京航空航天大学 | Rapid generating method of milling path of web of airplane structural member |
| CN103365243A (en) * | 2013-06-18 | 2013-10-23 | 南京航空航天大学 | Method for rapidly generating corner side milling process path |
| CN111014892A (en) * | 2019-12-13 | 2020-04-17 | 华中科技大学鄂州工业技术研究院 | Welding seam track monitoring system |
| CN111097664A (en) * | 2019-12-20 | 2020-05-05 | 广西柳州联耕科技有限公司 | Real-time deviation rectifying method based on robot gluing |
| CN111192307A (en) * | 2019-12-20 | 2020-05-22 | 广西柳州联耕科技有限公司 | Self-adaptive deviation rectifying method based on laser cutting of three-dimensional part |
| CN113172307A (en) * | 2021-03-24 | 2021-07-27 | 苏州奥天智能科技有限公司 | Industrial robot system of visual module based on laser and visible light fusion |
| CN113634964A (en) * | 2021-08-25 | 2021-11-12 | 武汉理工大学 | Gantry type robot welding equipment and welding process for large-sized component |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115971378A (en) * | 2023-03-16 | 2023-04-18 | 天津丰通晟源科技有限公司 | Special-shaped spring production control method and system based on visual detection |
| CN116518868A (en) * | 2023-07-05 | 2023-08-01 | 深圳市海塞姆科技有限公司 | Deformation measurement method, device, equipment and storage medium based on artificial intelligence |
| CN116518868B (en) * | 2023-07-05 | 2023-08-25 | 深圳市海塞姆科技有限公司 | Deformation measurement method, device, equipment and storage medium based on artificial intelligence |
| CN117146709A (en) * | 2023-10-30 | 2023-12-01 | 钛玛科（北京）工业科技有限公司 | Deviation rectifying control device and system based on automatic selection of sensor |
| CN117146709B (en) * | 2023-10-30 | 2024-02-02 | 钛玛科（北京）工业科技有限公司 | Deviation rectifying control device and system based on automatic selection of sensor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115586749B (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115586749B (en) | Workpiece machining track control method based on machine vision and related device | |
| CN111079602B (en) | Vehicle fine granularity identification method and device based on multi-scale regional feature constraint | |
| CN108510000B (en) | Method for detecting and identifying fine-grained attribute of pedestrian in complex scene | |
| CN111680542B (en) | Steel coil point cloud identification and classification method based on multi-scale feature extraction and Pointnet neural network | |
| WO2020192431A1 (en) | System and method for ordered representation and feature extraction for point clouds obtained by detection and ranging sensor | |
| CN113538486B (en) | Method for improving identification and positioning accuracy of automobile sheet metal workpiece | |
| CN111160407A (en) | Deep learning target detection method and system | |
| CN112085024A (en) | Tank surface character recognition method | |
| CN115032648B (en) | Three-dimensional target identification and positioning method based on laser radar dense point cloud | |
| CN116229189B (en) | Image processing method, device, equipment and storage medium based on fluorescence endoscope | |
| JP2018128897A (en) | Detection method and detection program for detecting attitude and the like of object | |
| CN115880953B (en) | Unmanned aerial vehicle management and control method and intelligent street lamp system | |
| CN115810133B (en) | Welding control method based on image processing and point cloud processing and related equipment | |
| CN113936210A (en) | Anti-collision method for tower crane | |
| CN115147745A (en) | Small target detection method based on urban unmanned aerial vehicle image | |
| CN110472640B (en) | Target detection model prediction frame processing method and device | |
| CN115797962A (en) | Wall column identification method and device based on assembly type building AI design | |
| JP6701057B2 (en) | Recognizer, program | |
| CN114663857A (en) | Point cloud target detection method and device and domain controller | |
| CN117237902B (en) | Robot character recognition system based on deep learning | |
| CN116266387A (en) | YOLOV4 image recognition algorithm and system based on re-parameterized residual error structure and coordinate attention mechanism | |
| CN113259883A (en) | Multi-source information fusion indoor positioning method for mobile phone user | |
| CN116664851A (en) | Automatic driving data extraction method based on artificial intelligence | |
| CN111126513B (en) | Universal object real-time learning and recognition system and learning and recognition method thereof | |
| CN115830342A (en) | Method and device for determining detection frame, storage medium and electronic device |
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 |