CN210006066U - part type recognition device for crankshaft flexible production line - Google Patents
part type recognition device for crankshaft flexible production line Download PDFInfo
- Publication number
- CN210006066U CN210006066U CN201921293972.9U CN201921293972U CN210006066U CN 210006066 U CN210006066 U CN 210006066U CN 201921293972 U CN201921293972 U CN 201921293972U CN 210006066 U CN210006066 U CN 210006066U
- Authority
- CN
- China
- Prior art keywords
- detection station
- production line
- communication system
- measuring head
- communication
- 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.)
- Expired - Fee Related
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 46
- 238000004891 communication Methods 0.000 claims abstract description 46
- 238000001514 detection method Methods 0.000 claims abstract description 38
- 238000011179 visual inspection Methods 0.000 claims abstract description 8
- 238000007689 inspection Methods 0.000 claims description 34
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000005622 photoelectricity Effects 0.000 abstract description 3
- 238000009434 installation Methods 0.000 abstract 1
- 238000000034 method Methods 0.000 description 15
- 238000012706 support-vector machine Methods 0.000 description 8
- 239000000523 sample Substances 0.000 description 7
- 238000003754 machining Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 230000006698 induction Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The utility model relates to a part kind recognition device for flexible production line of bent axle, the device include visual inspection gauge head, photoelectricity limit switch, communication system, robotic arm, the truss comprises three crossbeams and vertical beams, vertical beam below installation visual inspection gauge head, robotic arm installs on vertical beam, the bent axle part is placed on detection station, detection station includes detection station platform and two support V type grooves and two fixed V type grooves of symmetry setting on detection station platform, detection station platform's lower part is installed on the slide rail, two support V type grooves all are located the outside of two fixed V type grooves, visual inspection gauge head and communication system's communication mode is the bluetooth communication, and communication system and production line controller's communication adopt RS232 communication mode, the device acquires, handles, classifies and robotic arm is integrated to plays, and visual inspection gauge head directly handles data, satisfies the requirement of rapid processing on the production line.
Description
Technical Field
The utility model relates to a flexible production line part type discernment technical field specifically is kinds of part type recognition device that are used for the flexible production line of bent axle.
Background
The flexible production line is formed by arranging series high-speed composite machining centers and efficient special machine tools according to the process flow, and is connected with a manipulator truss by an automatic conveying device, the manipulator is responsible for quickly, efficiently and accurately grabbing parts and identifying the types of the parts by an identification device and conveying the parts to corresponding machining stations, the machine vision has the advantages of non-contact, high reliability, high automation degree and the like, and is widely applied to image identification of parts in the production line.
Patent No. ZL201210297994.9 discloses a workpiece recognition device on an automatic production line, wherein a cylinder connected with a pneumatic control device is mounted on a frame, a piston rod matched with the cylinder is connected with a moving plate, the moving plate is connected with a guide mechanism, a sliding sleeve is arranged on the moving plate, a spring is sleeved on a moving shaft matched with the sliding sleeve, an impact head is connected with end of the moving shaft, the spring presses the moving shaft to the side of the impact head , the other end of the moving shaft is connected with an induction head matched with a proximity switch, a plurality of groups of moving shafts are provided, after a workpiece is conveyed to a certain position by a roller type workpiece conveying device, the cylinder acts to drive the moving plate to move towards the workpiece, so that the impact head at the front end of the moving shaft is contacted with the workpiece, the pressed moving shaft moves relative to the sliding sleeve, the spring is compressed, the induction heads at the rear end of the moving shaft are mutually inducted with a proximity switch, a plurality of induction heads send out signals, a control system can recognize the workpiece.
SUMMERY OF THE UTILITY MODEL
To the not enough of prior art, the utility model discloses the technical problem who plans to solve is, provides kinds of part type recognition device that are used for the flexible production line of bent axle.
The utility model provides a technical scheme of the technical problem be, a part kind quick identification device for flexible production line of bent axle is provided, its characterized in that, the device includes inspection gauge head, photoelectricity limit switch, communication system, robotic arm, communication system is connected with the production line controller through the RS232 interface, the truss comprises three crossbeams and vertical beams, three crossbeams are the I shape, connect vertical beam on the crossbeam in the middle of the I shape, the inspection gauge head is installed to vertical beam below, robotic arm installs on vertical beam, robotic arm can move up and down along with the truss from side to side;
the vision inspection measuring head comprises an image acquisition module, an image processing board and a vision inspection measuring head power supply, wherein the vision inspection measuring head power supply supplies power to the whole vision inspection measuring head, the end of the vision inspection measuring head is fixed on a vertical beam of a truss through two screws, and the end of the vision inspection measuring head is provided with the image acquisition module which comprises a digital CCD camera, a telecentric lens and an LED annular light source;
the crankshaft part is placed on a detection station; the detection station comprises a detection station platform, and two supporting V-shaped grooves and two fixing V-shaped grooves which are symmetrically arranged on the detection station platform; the lower part of the detection station platform is arranged on the slide rail, and the detection station platform can move back and forth on the rail under the driving of the motor; the two supporting V-shaped grooves are positioned at the outer sides of the two fixed V-shaped grooves;
a photoelectric limit switch is arranged on a cross beam between the tracks right below the CCD camera and used for judging whether a detection station reaches a position or not;
the communication system comprises an ARM processor, a communication conversion chip and Bluetooth; the communication mode of the vision inspection measuring head and the communication system is Bluetooth communication, the communication of the communication system and the production line controller adopts an RS232 communication mode, and the ARM processor is electrically connected with the photoelectric limit switch.
The model of the CCD camera is MV-EM120M, and the photosensitive area of the CCD camera is 4.8mm multiplied by 3.6 mm; the model of the telecentric lens is HX 2X-T110; the telecentric lens is a zoom lens, the focal length is 12mm-36mm, and the minimum object distance is 20 cm; the LED annular light source is in the model number of HDR-90-45.
The supporting V-shaped groove is 10 cm higher than the fixed V-shaped groove.
A method for quickly identifying the types of loaded parts in a DSP.
Compared with the prior art, the utility model discloses beneficial effect lies in:
(1) this recognition device acquires, handles, categorised and robotic arm is integrated to plays, and the inspection gauge head is direct to be handled data, passes through bluetooth wireless transmission with communication system, then passes through communication system and production line controller transmission information, has reduced the wiring of production line, can satisfy the requirement of real-time, has improved the production efficiency of production line, is fit for the quick discernment of the work piece kind on the flexible production line.
(2) The device adopts the truss to connect all parts, and the adjustable height and the space position of the truss can be suitable for the detection of different types of crankshaft parts with different lengths and sizes.
Drawings
FIG. 1 is a schematic view of the overall structure of the device for identifying the type of a part in a flexible production line of a crankshaft according to the present invention;
FIG. 2 is a schematic view of the detection station of the part type identification device for the crankshaft flexible production line of the present invention;
FIG. 3 is a front view of the working state of the part type identification device for the flexible production line of the crankshaft of the present invention;
FIG. 4 is a side view of the working state of the part type identification device for the flexible production line of the crankshaft of the present invention;
fig. 5 is a schematic structural diagram of a vision inspection probe of the part type identification device for the crankshaft flexible production line according to the present invention;
FIG. 6 is a schematic diagram of the hardware connection of the vision inspection probe image processing board of the part type recognition device for the crankshaft flexible production line of the present invention;
fig. 7 is a schematic diagram of a hardware structure of a communication system of the device for identifying the type of a part on a flexible production line of a crankshaft according to the present invention;
FIG. 8 is a flow chart of the fast identification of the crankshaft part type based on deep learning of the part type identification device for the flexible production line of the crankshaft of the present invention; (in the figure, 1, a truss, 2, a mechanical arm, 3, a vision inspection probe, 4, a crankshaft part, 5, a detection station, 6, a sliding rail, 7, a photoelectric limit switch, 8, Bluetooth, 201, a support groove, 301, an LED annular light source, 302, a telecentric lens, 303, a digital CCD camera, 304, an image processing board, 305, a connecting piece, 306, a vision inspection probe power supply, 501, a detection station platform, 502, a support V-shaped groove, 503, a fixed V-shaped groove, 601, a beam between rails)
Detailed Description
Specific embodiments of the present invention are given below, and the specific embodiments are only used for describing the present invention in detail at step , and do not limit the protection scope of the present invention.
The utility model provides parts type quick identification device (refer to fig. 1-5) for flexible production line of bent axle, including visual inspection gauge head 3, photoelectricity limit switch 7, communication system, production line controller, robotic arm 2, communication system is connected with the production line controller through the RS232 interface, truss 1 comprises three crossbeams and vertical beams, three crossbeams are the I shape, connect the vertical beam on the crossbeam in the middle of the I shape, install visual inspection gauge head 3 below the vertical beam, the crossbeam in the middle drives robotic arm and moves about around under drive arrangement's drive, and the vertical beam can reciprocate, the structure of concrete drive arrangement can adopt prior art, like linear module;
the vision inspection measuring head 3 comprises an image acquisition module, an image processing board 304, a connecting piece 305 and a vision inspection measuring head power supply 306, the vision inspection measuring head power supply 306 supplies power to the whole vision inspection measuring head, the end of the vision inspection measuring head 3 is fixed on a vertical beam of a truss through two screws, the other end is provided with the image acquisition module, the image acquisition module comprises a digital CCD camera 303, a telecentric lens 302 and an LED annular light source 301, the digital CCD camera 303 is connected with the telecentric lens 302, an interface of the digital CCD camera 303 is connected with the image processing board 304, the image processing board adopts TMS320C6748DSP of TI company as a core processor, Spartan-6FPGA of Xilinx company is adopted to realize logic control of a circuit, the FPGA is hung with Bluetooth and gigabit interface chips, the hung with gigabit interface chips is used for completing communication with the camera, the FPGA is interconnected with the DSP through a parallel port for transmission of image data, the DSP and the FPGA sends related information through instructions, the hardware connection schematic diagram is shown in figure 6, the digital camera 303 of a gigabit Ethernet interface is selected, and the digital camera 303 of the crankshaft is connected with a vision inspection measuring head, and the part of a vision inspection device capable of detecting a vision inspection measuring head, wherein the crankshaft is connected with a serial port 634.
The model of the digital CCD camera 303 is MV-EM120M, and the resolution is 1280 x 960; the photosensitive area of the digital CCD camera 303 is 4.8mm multiplied by 3.6mm, and three image acquisition modes of continuous, external triggering and soft triggering are provided;
the model of the telecentric lens 302 is HX 2X-T110; the telecentric lens 302 is a zoom lens, the focal length is 12mm-36mm, and the minimum object distance is 20 cm;
the LED annular light source 301 is of the type HDR-90-45, the inner diameter is 45mm, the outer diameter is 90mm, and the power is 9W;
the crankshaft part 4 is placed on a detection station 5, the detection station 5 comprises a detection station platform 501, two supporting V-shaped grooves 502 and two fixing V-shaped grooves 503, the two supporting V-shaped grooves 502 and the two fixing V-shaped grooves 503 are symmetrically arranged on the detection station platform, the lower portion of the detection station platform is mounted on a sliding rail 6, the detection station platform can move back and forth on the rail under the driving of a motor, the two supporting V-shaped grooves 502 are located on the outer sides of the two fixing V-shaped grooves 503, the supporting V-shaped grooves 502 are about 10 cm higher than the fixing V-shaped grooves 503, the two fixing V-shaped grooves 503 inside the supporting V-shaped grooves can move, two circular holes are formed in the detection station platform every 10 cm, the V-shaped grooves are fixed on the detection station platform through the circular holes and screws, the distance between the middle fixing V-shaped grooves is used for placing crankshaft parts with the shortest length, when the crankshaft is longer, the two sides of the central shaft of the crankshaft can be placed on the supporting V-shaped grooves 502, the middle crank portion can be supported by the fixing V-shaped grooves 503, the stress in the middle of the crankshaft can be reduced, the crankshaft can be placed more.
The lower part of the detection station 5 is arranged on a slide rail, the detection station 5 can move back and forth on the slide rail under the drive of a motor, a cross beam 601 between the rails is used for enabling the rails to be more stable, a photoelectric limit switch is arranged on the cross beam 601 between the rails right below the CCD camera, and the photoelectric limit switch 7 is used for judging whether the detection station 5 reaches the position;
the communication system comprises an STM32L0 low-power-consumption series ARM processor, a communication conversion chip MAX3232 and Bluetooth; the communication mode of a vision inspection measuring head (DSP) and a communication system is Bluetooth communication, the communication of the communication system and a production line controller adopts an RS232 communication mode, an ARM processor preferably selects STM32L051C8T6, the ARM processor is electrically connected with a photoelectric limit switch, and the hardware connection relationship is shown in figure 7;
the production line controller controls the whole crankshaft flexible production line; the production line controller is of an existing structure.
The mechanical arm 2 is used for conveying parts to corresponding machining stations or removing unqualified crankshaft parts from a production line, the mechanical arm end is fixed on a vertical beam of the truss, and the other end end grabs the parts to be identified.
The utility model discloses a part kind recognition device theory of operation and workflow for flexible production line of bent axle are:
the principle is that after a detection station of a crankshaft part reaches a designated position, a photoelectric limit switch 7 sends signals to a production line controller through a communication system, the production line controller sends an instruction to a vision inspection head through the communication system after receiving the signals, the vision inspection head receives an acquisition instruction, frames of images are acquired, a parallel port receiving program in a DSP is started at the same time, the FPGA enters an acquisition state, digital images acquired by a CCD are converted into parallel data through a gigabit network and then are subjected to simple preprocessing (noise filtering, illumination balancing and the like), the image data subjected to the simple preprocessing by the FPGA are transmitted into the DSP through a parallel interface, the DSP processes the images (the identification method of the application is loaded in the DSP) and extracts the characteristics, the extracted characteristic values are input into a support vector machine classifier which is designed and trained to classify related characteristics, the classification results (the crankshafts are types of crankshafts, the crankshafts are required to be placed on different processing stations and call corresponding processing programs according to different crankshafts, the processing modes of each crankshaft are different, the cutting parameters and the logical values are transmitted to the FPGA through a serial port control system, and the FPGA controls the processing system to send the processing steps according to a working procedure of the FPGA.
The utility model discloses identification means's working process includes following step:
the process of the method for quickly identifying the types of the crankshaft parts comprises the following steps: removing the background of the image, extracting a workpiece part, and subtracting the image with the crankshaft part from the image without the crankshaft part which is shot in advance by adopting an image subtraction algorithm; respectively extracting the features of the image by using a skeleton method and an AlexNet network; fusing the two extracted features; inputting the fused features into a trained support vector machine for classification and identification; the method comprises the following specific steps:
(1) subtracting the image with the crankshaft part from the image without the crankshaft part in advance by adopting an image subtraction algorithm, and extracting the workpiece part;
(2) carrying out corrosion, expansion, opening operation and closing operation on the extracted workpiece image to obtain a gray image;
(3) selecting 1000 gray images as a gray image library for the sample to train the network and the support vector machine; extracting skeleton characteristics and training a convolutional neural network by 1000 training samples respectively;
carrying out skeleton extraction:
① Euclidean distance transform and projective transform of binary image (gray image):
binary images I are X ∪ S, X is foreground shape, S is background, Euclidean distance transformation D of the images IsFor every points X (X) on the foreground shape X1,x2) Shortest distance to background S:
wherein,
y being a point on the background S, i.e. y1Corresponds to x1，y2Corresponds to x2。
Projective transformation:
PS(x)＝{y∈S|d(x,y)＝DS(x)}
② calculating deltax
δxCan be seen as the radius of maximum shape deformation at the point where the backbone branch connected to x is propped up. Order:
Nx＝{μ∈I|d(x,μ)＝1}
Nxis the immediate neighbor of point x on the image. First, the Euclidean distance transform D of a point x and its adjacent points mu on an image is calculateds(x) And Ds(mu) finding projection points
And
connection pointAnd
obtain a line segment Zx,μZ is Zx,μTo reduce the amount of computation by step , point Z is approximated as line segment Zx,μThe midpoint of (a). DzThe Euclidean distance from the point z to the background S, and the radius of the maximum circle which is centered at the point z and does not intersect with the background S is rx,μCalculated from the following equation:
rx,μ＝sup{Dz|z∈Zx,μ}
let m ═ x + μ)/2 be the line segment [ x, μ%]So that δ can be calculatedx：
Condition
Can avoid the adjacent point mu to x off-line segment
The upper perpendicular bisector is close.
③ obtaining skeleton map
By satisfying the condition deltaxThe series skeleton points of being equal to or larger than delta form a skeleton map MA of the foreground shape X, and the formula of the skeleton map MA is as follows:
MAδ(x)＝{x∈X|δx≥δ}
where δ is parameters for eliminating spurious branches.
The skeleton map is represented by a skeleton feature matrix as a × b matrixes, wherein a and b are integers greater than zero.
Extracting gray level image features:
the embodiment of the utility model provides a selected the basis that the classic model AlexNet of deep learning discerned as convolutional neural network, AlexNet model has added Dropout layer behind the full tie layer and has prevented the overfitting, has used MAXPooling, make the pooling layer overlap, can alleviate the overfitting, used nonlinear activation function ReLU, compare with traditional activation function sigmoid, tanh, ReLU mathematical formula is simple, only threshold values 0, the computational rate is fast, no phenomenon of gradient dispersion takes place, and when using random gradient descent method to optimize the network, the convergence is faster, the mathematical expression of ReLU function is:
in the embodiment, the size of an input image, the number of convolutional layers and the size of a convolutional kernel need to be adjusted according to the characteristics of a crankshaft image, in the other aspect, the network is optimized, wherein is to remove LRN (local response regression ) layers in an AlexNet network model to reduce the calculation time and the memory consumption, in the second aspect, the number of Dropout layers is reduced, only Dropout layers are used in the network, because Dropout can increase the iteration times required by convergence, and the Dropout layers are added into the last fully-connected layers, and the principle of Dropout is that in each training process, the Dropout algorithm can randomly make the output values of part of neuron nodes 0 with the probability determined by , namely randomly abandons part of neurons to prevent complex synergistic effect on training data, so that the networking capability is improved;
on the basis of ensuring the detail information of an input image, in order to reduce the data volume, converting the acquired crankshaft image into an image with the size of 150 multiplied by 150 to be used as the input of a network;
the method adopts 3 convolution layers and 2 full-connection layers, the convolution kernel size of th convolution layer is 3 multiplied by 3, the second convolution kernel size is 7 multiplied by 7, the third convolution kernel size is 13 multiplied by 13, each convolution kernel should have nonlinear activation functions ReLU, and the number of parameters can be reduced;
training an AlexNet network through 1000 training samples, and after the network is trained, extracting the characteristics of the gray level image through the network, wherein the extracted characteristics are b-dimensional vectors with the same column as the skeleton characteristic matrix;
fusing the two features to obtain (a +1) xb dimensional matrixes, inputting the fused features into a support vector machine to train the support vector machine, and downloading the support vector machine and an AlexNet network into an image processing board for identifying the crankshaft on the production line after the support vector machine and the AlexNet network are trained;
and 4, inputting the digital image preprocessed in the step 3 into a crankshaft part type rapid identification method, performing background removal operation on the image, performing skeleton feature extraction in the skeleton extraction mode, performing gray image feature extraction on the AlexNet network, performing two feature fusion, inputting the fused features into a trained support vector machine for classification, outputting a part classification result, sending the classification result to a production line controller through a communication system, and enabling the production line controller to call corresponding machining programs according to machining procedures under the control of the information or to remove parts on a detection station according to instructions of a mechanical arm.
The crankshaft part is identified through feature fusion, and the identification rate can reach more than 95%.
The utility model discloses the nothing is described the part and is used in prior art.
Claims (3)
- The part type identification device for the crankshaft flexible production line is characterized by comprising a visual inspection measuring head, a photoelectric limit switch, a communication system and a mechanical arm, wherein the communication system is connected with a production line controller through an RS232 interface;the vision inspection measuring head comprises an image acquisition module, an image processing board and a vision inspection measuring head power supply, wherein the vision inspection measuring head power supply supplies power to the whole vision inspection measuring head, the end of the vision inspection measuring head is fixed on a vertical beam of a truss through two screws, and the end of the vision inspection measuring head is provided with the image acquisition module which comprises a digital CCD camera, a telecentric lens and an LED annular light source;the crankshaft part is placed on a detection station; the detection station comprises a detection station platform, and two supporting V-shaped grooves and two fixing V-shaped grooves which are symmetrically arranged on the detection station platform; the lower part of the detection station platform is arranged on the slide rail; the two supporting V-shaped grooves are positioned at the outer sides of the two fixed V-shaped grooves;a photoelectric limit switch is arranged on a beam between the tracks right below the CCD camera;the communication system comprises an ARM processor, a communication conversion chip and Bluetooth; the communication mode of the vision inspection measuring head and the communication system is Bluetooth communication, the communication of the communication system and the production line controller adopts an RS232 communication mode, and the ARM processor is electrically connected with the photoelectric limit switch.
- 2. The part kind recognition device according to claim 1, wherein the CCD camera model is MV-EM120M, the CCD camera light sensing area is 4.8mm x 3.6 mm; the model of the telecentric lens is HX 2X-T110; the telecentric lens is a zoom lens, the focal length is 12mm-36mm, and the minimum object distance is 20 cm; the LED annular light source is in the model number of HDR-90-45.
- 3. The part kind identification device of claim 1, wherein the supporting V-groove is 10 cm higher than the fixed V-groove.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201921293972.9U CN210006066U (en) | 2019-08-12 | 2019-08-12 | part type recognition device for crankshaft flexible production line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201921293972.9U CN210006066U (en) | 2019-08-12 | 2019-08-12 | part type recognition device for crankshaft flexible production line |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN210006066U true CN210006066U (en) | 2020-01-31 |
Family
ID=69311288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201921293972.9U Expired - Fee Related CN210006066U (en) | 2019-08-12 | 2019-08-12 | part type recognition device for crankshaft flexible production line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN210006066U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363254A (en) * | 2019-08-12 | 2019-10-22 | 河北工业大学 | Part category method for quickly identifying and identification device for crankshaft flexible production line |
-
2019
- 2019-08-12 CN CN201921293972.9U patent/CN210006066U/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363254A (en) * | 2019-08-12 | 2019-10-22 | 河北工业大学 | Part category method for quickly identifying and identification device for crankshaft flexible production line |
| CN110363254B (en) * | 2019-08-12 | 2024-02-02 | 河北工业大学 | Quick identification device for types of parts of flexible crankshaft production line |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110363254B (en) | Quick identification device for types of parts of flexible crankshaft production line | |
| CN109772724B (en) | Flexible detection and analysis system for major surface and internal defects of castings | |
| CN104156726B (en) | A kind of workpiece identification method and device based on geometric characteristic | |
| CN106853639A (en) | A kind of battery of mobile phone automatic assembly system and its control method | |
| CN111266304A (en) | Coal briquette identification, detection and sorting system | |
| CN111046948A (en) | Point cloud simulation and deep learning workpiece pose identification and robot feeding method | |
| CN113383227A (en) | Defect inspection device | |
| CN210006066U (en) | part type recognition device for crankshaft flexible production line | |
| CN113145492A (en) | Visual grading method and grading production line for pear appearance quality | |
| CN110524697B (en) | Automatic glaze spraying system for toilet bowl blank and positioning method thereof | |
| CN116934719B (en) | Automatic detection system for belt conveyor | |
| CN112718552A (en) | Automatic quality defect inspection device for LED circuit board and working method thereof | |
| CN116337887A (en) | Method and system for detecting defects on upper surface of casting cylinder body | |
| CN113012228B (en) | Workpiece positioning system and workpiece positioning method based on deep learning | |
| CN214682976U (en) | Garbage recognition and automatic sorting device | |
| CN112326552A (en) | Tunnel block falling disease detection method and system based on vision and force perception | |
| CN117269168A (en) | New energy automobile precision part surface defect detection device and detection method | |
| CN105954288A (en) | Bare board detection and sorting system and detection method used after electrolytic manganese negative plate stripping | |
| CN114800494A (en) | Box moving manipulator based on monocular vision | |
| TWI838236B (en) | System and method for personnel detection | |
| CN111842212A (en) | Automatic feeding structure, appearance detection device, mask appearance detection device and method | |
| KR102677763B1 (en) | Low quality Garlic sorting system | |
| KR102621770B1 (en) | Garlic sorting system | |
| CN212883587U (en) | Coal briquette identification, detection and sorting system | |
| CN116152228B (en) | Tire defect detection method and system based on machine vision and machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200131 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |