WO2020175308A1 - 切屑検出装置、工作機械、切屑検出方法、学習用画像合成装置 - Google Patents
切屑検出装置、工作機械、切屑検出方法、学習用画像合成装置 Download PDFInfo
- Publication number
- WO2020175308A1 WO2020175308A1 PCT/JP2020/006748 JP2020006748W WO2020175308A1 WO 2020175308 A1 WO2020175308 A1 WO 2020175308A1 JP 2020006748 W JP2020006748 W JP 2020006748W WO 2020175308 A1 WO2020175308 A1 WO 2020175308A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- mesh
- image
- chip
- chips
- cleaning
- 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.)
- Ceased
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
- B23Q17/2409—Arrangements for indirect observation of the working space using image recording means, e.g. a camera
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q11/00—Accessories fitted to machine tools for keeping tools or parts of the machine in good working condition or for cooling work; Safety devices specially combined with or arranged in, or specially adapted for use in connection with, machine tools
- B23Q11/0042—Devices for removing chips
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30141—Printed circuit board [PCB]
Definitions
- Chip detection device machine tool, chip detection method, learning image synthesizer
- the present invention relates to a chip detection device for detecting chips generated when a work is machined by a machine tool, a machine tool, a chip detection method, a learning image synthesizing device, and the like.
- Patent Document 1 Japanese Patent Application Laid-Open No. 7-1 08 4 3 5
- the chips generated during processing have innumerable variations in shape, color, size, and assembly status, etc., depending on the processing conditions, tools used, work materials, and so on.
- the environment inside the machine which is the background of the image, has conditions such as structure and position, types of workpieces and jigs, presence/absence of coolant (cleaning liquid), and illuminance, and there are various patterns of combinations.
- the present invention provides a chip detecting device, a machine tool, a chip detecting method, a learning image synthesizing device, and the like.
- a chip detection device is produced when a workpiece is machined by a machine tool in order to solve the problem of simply and accurately detecting the chips produced when the workpiece is machined by the machine tool.
- a chip detection device for detecting chips comprising a mesh division unit for dividing at least a part of an area image of the target area for detecting chips into a plurality of mesh images with a predetermined mesh size, and the chip.
- a chip information determination unit that determines the chip information for each of the mesh images by using a determination parameter set in advance for determining the chip information regarding
- chips generated when a workpiece is machined by a machine tool can be easily detected.
- Fig. 1 is a block diagram showing an embodiment of a chip detection device and a machine tool including the same.
- FIG. 2 A diagram showing a state in which an area image is divided into a plurality of mesh images in the present embodiment.
- FIG. 3 A diagram showing how an area image is constructed by combining a plurality of partial images in the present embodiment.
- FIG. 4 is a diagram showing a division process by the mesh division unit of the present embodiment.
- FIG. 5 A diagram for explaining a “class” which is chip information of the present embodiment.
- FIG. 6 is a diagram showing a determination process by the chip determination model of the present embodiment.
- FIG. 6 is a diagram showing an example of a feature amount of a mesh image.
- FIG. 8 is an example of a display screen showing (3) a determination result before correction and (a) a determination result after correction in the teaching mode of the present embodiment.
- FIG. 9 is a flow _ Chiya _ Bok showing chips detecting method performed by the chip detecting device of the present embodiment.
- FIG. 10 A flow chart showing a chip detection process in the present embodiment.
- FIG. 11 A flow chart showing the teaching process in the present embodiment.
- FIG. 12 is a block diagram showing the configuration of a learning image synthesizing device in the present embodiment.
- the chip detecting device 1 of the present embodiment detects chips generated when a work is machined by the machine tool 10. Specifically, as shown in FIG. 1, the chip detecting device 1 controls the photographing means 11 provided on the machine tool 10 to acquire an image of a target area for detecting chips. Then, as shown in FIG. 2, the image is divided into a plurality of mesh images, and chip information regarding chips is determined for each of the mesh images.
- FIG. 1 the chip detecting device 1 controls the photographing means 11 provided on the machine tool 10 to acquire an image of a target area for detecting chips.
- the image is divided into a plurality of mesh images, and chip information regarding chips is determined for each of the mesh images.
- chip information is a concept that includes information that can be detected regarding chips, such as the presence or absence of chips, the amount of chips, the shape of chips, the type of chips, or a combination thereof. Therefore, in the present embodiment, as will be described later, the “class” indicating the ease of cleaning the chips is determined as the chip information.
- the machine tool 10 is a machine for performing machining such as cutting and grinding on a workpiece such as metal, wood, stone, and resin.
- the machine tool 10 is numerically controlled according to a drive signal output from a numerical control device as the chip detection device 1.
- the chip detection device 1 and the machine tool ⁇ 0 2020/175308 4 ⁇ (: 171? 2020 /006748
- the numerical control device of the machine is composed of the same device, but is not limited to this.
- a computer other than the machine tool may be used as the chip detection device 1.
- the machine tool 10 illuminates the inside of the machine with a photographing means 11 for photographing the inside of the machine, a table 12 on which a work is placed, a cleaning means 13 for cleaning chips. Lighting means 14 and.
- the photographing means 11 is for photographing the target area for detecting chips.
- the photographing means 11 is an automatic tool changer (8 1 ⁇ ⁇ 11131
- the image capturing means 11 does not need to be installed in the machine, and is removable from the spindle of the machine tool in consideration of the merit that the exposure time to the coolant mist is relatively short.
- I'm using a camera Since the camera can be attached to the spindle of the machine tool, if the spindle of the machine tool is moved, the camera can be moved to follow it. Therefore, the shooting range and shooting direction of the camera can be easily changed.
- the 8 camera is described as an example because the shooting range and shooting direction of the camera can be easily changed, but the present invention is not limited to this configuration.
- One or more fixed cameras may be installed on board. With this configuration, the shooting time is relatively short, and the advantage is that you can shoot even during processing.
- the angle of view of the photographing means 11 is smaller than the target area, as shown in Fig. 3, if the camera is a straight camera, a plurality of partial images are photographed while moving the spindle, and a fixed camera may be used. For example, multiple fixed images are taken with multiple fixed cameras. Then, the partial area images covering the entire target area may be configured by combining the partial images.
- the cleaning means 13 cleans and removes the chips generated by machining the workpiece. ⁇ 0 2020/175 308 5 ⁇ (: 171? 2020 /006748
- the cleaning means 13 is composed of a programmable nozzle capable of controlling the injection direction of the cleaning liquid. Then, according to a cleaning control signal from a cleaning processing unit 48, which will be described later, a cleaning liquid such as a coolant liquid is sprayed onto a desired area on the work or the table 12.
- the programmable nozzle is used as the cleaning unit 13, but the present invention is not limited to this configuration.
- a plurality of cleaning nozzles having a fixed spraying direction may be installed so as to cover the entire target area, and the cleaning liquid may be sprayed from the cleaning nozzle near the area where cleaning is required.
- the illumination means 14 illuminates the inside of the machine so that chips can be easily detected.
- the illuminating means 14 is fixed to the inside of the machine and is constituted by a programmable light capable of controlling on/off switching, brightness, irradiation direction and the like.
- the configuration is not limited to this, and the light attached to the 8-chome camera may be used.
- the illumination means 14 may be rotated or a polarization filter may be attached to the camera lens of the imaging means 11.
- the chip detecting device 1 is composed of a computer such as a numerical control device for controlling the machine tool 10. As shown in Fig. 1, mainly, the user inputs instructions and displays judgment results.
- each constituent means will be described in detail.
- the chip detection device may be a device different from the numerical control device for controlling the machine tool 10.
- the chip detector is designed to use at least a part of the area image of the target area where chips are to be detected. ⁇ 0 2020/175308 6 ⁇ (: 171? 2020 /006748
- the chip information judgment for judging the chip information for each of the mesh images is performed.
- the display input means 2 is composed of a touch panel and has both an input function of receiving an instruction input from a user and a display function of displaying a chip determination result or the like.
- the touch panel is used as the display input means 2, but the present invention is not limited to this configuration, and a display means having only a display function and an input means having only an input function may be used. You may have each separately.
- the storage means 3 is a hard disk
- a program memory 3 1 a judgment parameter memory 3 2 and a judgment result memory. It has a unit 33, a cleaning condition storage unit 34, and a cleaning frequency storage unit 35.
- a chip detection program 13 for controlling the chip detection device 1 of the present embodiment is installed in the program storage unit 31. Then, the arithmetic processing means 4 executes the chip detection program 13 to cause the chip detection device 1 as the computer to function as each component described later.
- the usage form of the chip detection program 13 is not limited to the above configuration.
- the chip detection program 13 may be stored in a non-temporary recording medium that can be read by a computer and directly read from the recording medium and executed.
- it may be used from an external server or the like in the cloud computing method or the 8 3 (8 ⁇ les 1081; ⁇ ⁇ ⁇ $ 61 ⁇ _ 106 ⁇ "0 _1 ⁇ 16") method.
- the determination parameter storage unit 32 stores a determination parameter preset for determining chip information regarding chips. This embodiment ⁇ 0 2020/175308 7 ⁇ (: 171? 2020 /006748
- a chip determination model that applies learning parameters to a learning/reasoning model, which is a machine learning algorithm, is used as an algorithm for determining chip information. For this reason, learning parameters obtained by pre-learning are stored as the determination parameters by using teacher data that is a set of mesh images and chip information.
- the judgment parameter storage unit 32 stores learning parameters as judgment parameters for each of a plurality of different mesh sizes.
- the determination parameter storage unit 32 is additionally set with hyperparameters that left and right characteristics such as learning efficiency of the chip determination model.
- the determination result storage unit 33 stores the chip information determined for each mesh image as a determination result.
- the chip information determination unit 46 which will be described later, determines the probability of being the class as chip information for each of the mesh images for each class indicating the ease of cleaning the chips.
- the determination result is not limited to the probability for each class, and the class with the highest probability may be simply stored as the determination result. Further, as the number of classes increases, the determination becomes more difficult, and the performance becomes difficult to improve. Therefore, it is preferable to set the number to about several. Examples of the simplest classes are two classes, "Class 0" with no chips and "Class 1" with chips.
- the cleaning condition storage unit 34 is a cleaning condition for determining whether or not chips need cleaning. ⁇ 0 2020/175308 8 ⁇ (: 171? 2020 /006748
- a condition is set that combines the probability for each class and the position of the mesh area corresponding to each mesh image. For example, when the above-mentioned three classes are set, an efficient and high-performance judgment algorithm is realized by evaluating the following cleaning conditions (1) to (4) in descending order.
- the cleaning condition is set by combining the chip information and the position of the mesh area.
- the cleaning condition is not limited to this, and may be set using the chip information. Good. For example, if the above-mentioned two most simplified classes are set, regardless of the position of the mesh area, Class 0 (without chips) will not be washed and Class 1 (with chips) will not be washed. You may set the washing condition of washing.
- the cleaning number storage unit 35 stores the number of times the chips have been continuously cleaned.
- the cleaning frequency storage unit 35 stores the number of continuous cleanings, which is the number of continuous cleanings in the target area, and the number of local continuous cleanings, which is the number of continuous cleanings of the same mesh area. Is remembered. Then, by using these cleaning times, as will be described later, the automatic cleaning operation is prevented from never ending.
- the arithmetic processing means 4 As shown in FIG. 1, by executing the chip detection program 13 installed in the storage means 3, the target area setting section 41, the drive control section 42, Image data acquisition unit 43, mesh size selection unit 44, mesh division unit 45, chip information determination unit 46, cleaning necessity determination unit 47, cleaning processing unit 48, teaching process It functions as part 49 and additional learning part 50.
- the chip detection program 13 installed in the storage means 3
- the target area setting section 41 the drive control section 42
- Image data acquisition unit 43
- mesh size selection unit 44 mesh division unit 45
- chip information determination unit 46 chip information determination unit 46
- cleaning necessity determination unit 47 cleaning processing unit 48
- teaching process It functions as part 49 and additional learning part 50.
- the target area setting unit 41 sets a target area in which chips are to be detected.
- the target area setting unit 41 is responsive to the user's instruction input using the display input means 2, the entire inside of the machine, the table 12, the protector, the work periphery, a part of these, or these. Any target area is set, such as a combination of.
- the target area setting unit 41 sets the target area manually by the user, but the present invention is not limited to this configuration and may be set automatically. Good. Specifically, the target area setting unit 41 may acquire the shape of the work in each processing step by referring to the processing program and automatically set the target area so that the entire work is included. The target area may be set automatically only on the route.
- the drive control unit 42 drives and controls the motor of each axis provided in the machine tool 10.
- the drive control unit 42 is configured to generate a drive control signal for causing the machine tool 10 to execute cutting processing or the like based on the processing program, and output the drive control signal to the motor of each axis.
- the drive control unit 42 uses an automatic tool changer (e.g., 111; 0111131; ⁇ (; 1 " x 1 ⁇ 18 hits 6": 8 x x ) Is controlled so that the tool can be replaced with an eight camera.
- the image data acquisition unit 43 acquires an image of the entire target area.
- the image data acquisition unit 43 outputs a shooting control signal to the shooting unit 11 at a preset shooting timing and acquires an image of the entire area. Further, when the image capturing means 11 acquires a partial image smaller than the target area, a plurality of partial images are captured so as to cover the target area, and they are combined to acquire an image of the entire area.
- the photographing timing may be manually set by the user or may be automatically set in accordance with predetermined setting conditions.
- the setting conditions are, for example, between processes, when changing tools, at certain time intervals, and when a predetermined machining amount is reached.
- the machining amount can be estimated from the processing program. ⁇ 0 2020/175 308 10 ⁇ (: 17 2020 /006748
- the mesh size selection unit 44 selects the mesh size when dividing the image of the entire area into a plurality of mesh images.
- the mesh size selection unit 44 is configured to select one or a plurality of mesh sizes as candidates from preset mesh sizes based on a predetermined criterion.
- the present invention simplifies and increases the accuracy of chip detection by dividing the image of the entire area into mesh images. Therefore, the mesh size is an important factor that affects the accuracy of chip detection, and it is preferable to consider the following characteristics of chips when selecting the mesh size.
- the mesh size selection unit 44 has a mesh size such that the shape of the chips can be recognized based on the sizes of the chips, the work, and the jig, and It is preferable to select a mesh size such that the edge shape of the background work or jig is simplified. By selecting such a mesh size, the edge shape of the work or jig can be easily distinguished from the chips, and false detection can be reduced.
- the mesh size selection unit 44 is not limited to the above configuration, and the user may manually select from a plurality of preset mesh sizes.
- the mesh support ⁇ 0 2020/175308 1 1 ⁇ (: 171? 2020 /006748
- the mesh division unit 45 divides the area image into a plurality of mesh images.
- the mesh division unit 45 generates a plurality of mesh images by dividing the area image acquired by the image data acquisition unit 43 with the mesh size selected by the mesh size selection unit 44. It is designed to be completed.
- a grid-like or lattice-like mesh is adopted, but the shape is not limited to this, and a rhombus, a triangle, a honeycomb shape or the like may be used. Good.
- the entire area image is divided into mesh images, but the present invention is not limited to this, and areas where chip detection is unnecessary are not divided, and at least a part of the area image is divided. You may make it divide only.
- the captured image of the area and the mesh information in a grid pattern may be stored separately in association with each other.
- the mesh division unit 45 is not limited to generating individual mesh images, but superimposes the captured image and the lattice information and processes the regions of the captured image divided by the lattice as mesh regions. Good.
- the mesh area in this case corresponds to the mesh image described above.
- the mesh division unit 45 when the image data acquisition unit 43 acquires the area image by combining a plurality of partial images, the mesh division unit 45 performs the division process as shown in FIG. , Is designed to perform one of the following processes
- the chip information determining unit 46 determines chip information for each mesh image.
- the chip information determining unit 46 determines the chip information for each of the mesh images divided by the mesh dividing unit 45 using the determination parameters read from the determination parameter storage unit 32. Then, the judgment result storage section 33 is adapted to save the data.
- the chip information determination unit 46 is configured to determine, as the chip information, the “class” indicating the ease of cleaning the chips. This class is determined by comprehensively considering the amount of chips, density, size, length and shape. Specifically, as shown in Figure 5, there are few chips, scattered, small, short, difficult to get caught, etc. It is classified as a higher class when it is hard to wash, such as large, long, easy to get caught, etc.
- the chip information determining unit 46 is configured by a chip determining model in which learning parameters are applied to a learning/inference model that is a machine learning algorithm.
- a learning/inference model that is a machine learning algorithm.
- CNN convolutional neural network
- the chip information determination unit 46 When chip information is determined using the chip determination model as described above, the chip information determination unit 46 first performs noise removal processing, image size conversion processing, etc. on each mesh image. Pre-processing to improve the judgment accuracy is performed. Next, as shown in Fig. 6, the chip information judging unit 46 inputs the mesh image to the learned chip judgment model to which the learning parameters prepared in advance in the judgment parameter storage unit 32 are applied. The probability of each class is output as chip information for each mesh image.
- the convolutional neural network is adopted as the machine learning algorithm, but the machine learning algorithm is not limited to this, and another machine learning algorithm may be used.
- a machine learning algorithm that uses input data that is not an image, such as a support vector machine (SVM)
- SVM support vector machine
- filter processing is performed on the mesh image to The shape feature amount of the mesh image is calculated.
- the class may be determined for each mesh image by inputting the shape feature amount into the learned chip determination model.
- the machine learning algorithm is adopted as the chip information determining unit 46, but the present invention is not limited to this, and a non-machine learning algorithm such as a deterministic algorithm is used. May be.
- a mesh image tends to be more complicated as the number of chips increases, and the number of high image frequency components tends to increase. Therefore, the chip information determination unit 46 performs frequency analysis such as Fast Fourier Transform (FFT), as shown in Fig. 7, and uses the vector statistic of each mesh image as a feature amount. calculate. Then, the class as the chip information may be determined for each mesh image based on the magnitude relation between the feature amount and the threshold value set as a determination parameter in advance.
- FFT Fast Fourier Transform
- the fast Fourier transform has a high environmental dependency, and if small screw holes other than chips, coolant splashes, etc. are imaged in the mesh image, the high-frequency component increases and erroneous determination easily occurs. .. For this reason, an area image without chips is prepared in advance, the difference image between this area image and the area image to be judged is divided into meshes, and fast Fourier transform is performed to eliminate environment-dependent components. You may do so.
- the chip information determining unit 46 determines the chip information using a single algorithm, but the present invention is not limited to this configuration, and the chip information is determined by a plurality of algorithms. You may make a comprehensive determination based on the result.
- the chip information determination unit 46 performs the determination for all mesh sizes. ⁇ 0 2020/175308 14 ⁇ (: 171? 2020 /006748
- the chip information for each mesh size is stored in the determination result storage unit 33 as the determination result.
- the cleaning necessity determination unit 47 determines whether or not the cleaning of chips is necessary for each mesh image. In the present embodiment, the cleaning necessity determination unit 47 compares the cleaning conditions stored in the cleaning condition storage unit 34 with the chip information of each mesh image stored in the determination result storage unit 33. .. Then, it is determined that cleaning is required for mesh images that satisfy the cleaning conditions.
- the cleaning processing unit 48 controls the cleaning unit 13 to clean the chips.
- the cleaning processing unit 48 calculates the position of the mesh area (actual position in the machine tool 10) corresponding to the mesh image determined to require cleaning and directs it toward the mesh area.
- the cleaning liquid is sprayed from the cleaning means 13.
- the position of the mesh area can be specified, for example, by making an image memory (not shown) that stores the area image correspond to the coordinate system in the machine tool 10. Further, in the present embodiment, since the cleaning means 13 is composed of the programmable nozzle, the coolant is sprayed to all the mesh areas determined to require cleaning while changing the spraying direction.
- the cleaning processing unit 48 causes the cleaning frequency storage unit 35 to store the number of continuous cleanings performed continuously in the target area every time the cleaning operation is performed. , And the same mesh area are continuously washed and the local continuous washing frequency is updated. Then, the cleaning processing unit 48 carries out cleaning when the number of continuous cleanings and the number of local continuous cleanings are less than a predetermined threshold. On the other hand, when at least one of the number of continuous cleanings and the number of local continuous cleanings reaches a predetermined threshold value, a predetermined abnormal condition process is executed to avoid the situation where the automatic cleaning operation does not end indefinitely.
- the abnormal-time process is not particularly limited as long as it is a process for avoiding a situation where the automatic cleaning operation does not end indefinitely.
- the threshold of the number of continuous cleanings and the threshold of the number of local continuous cleanings are configured to be individually set by the user.
- a method of notifying the alarm for example, a method of outputting a warning sound from a speaker (not shown) may be mentioned, but the method is not limited to this as long as it can notify the user. .. Specifically, a message may be displayed on the display input means 2, and a light (not shown) may be turned on and off. If there are no users around the machine tool 10, an error message may be sent to the user's mobile terminal.
- the teaching processing section 49 confirms and corrects the judgment result by the chip information judging section 46.
- the teaching processing unit 49 first reads the determination result of each mesh image from the determination result storage unit 33 and displays it on the display input means 2 as shown in FIG. 8(a). Specifically, different determination results are colored differently, and each mesh image is displayed in an identifiable manner.
- the chip information determination unit 46 calculates the determination probability for each mesh image, and the determination result may be incorrect for mesh images in which the determination probability of any class is not so high. It may be highlighted as something that is more likely to be correct (no confidence in the judgment result) to assist the user in making corrections. For example, in the example shown in Fig. 8 (a), the classification probability of class 0 (without chips) is 0, and the classification probability of class 1 (with chips) is 1, the following four groups are classified.
- [0065] 1 is ⁇ to 30% (0 is 70 to 100%): “Confidence” of class 0 (no chips)
- the teaching processing unit 49 determines that The judgment result is acquired, and the judgment result stored in the judgment result storage unit 33 is corrected. Then, as shown in FIG. 8 (13), the corrected determination result is displayed on the display input means 2.
- the additional learning unit 50 performs additional learning using new teacher data.
- the additional learning unit 50 performs additional learning on the chip determination model with the teacher data consisting of the set of the determination result (correct chip information) corrected by the teaching processing unit 49 and the mesh image.
- the learning parameter that reflects the correct judgment result is acquired.
- the learning parameter is used as a new judgment parameter to update the judgment parameter in the judgment parameter storage unit 32.
- the additional learning unit 50 is provided in the oral chip detecting device 1 to execute the additional learning in the background, but the present invention is not limited to this configuration. That is, the learning data may be uploaded to a learning server on the network for additional learning, and the obtained learning parameter may be downloaded and used as a new determination parameter.
- an automatic cleaning mode and a teaching mode are prepared as the operation modes of the cleaning processing by the chip detection device 1, and either operation mode can be set.
- the automatic cleaning mode is a mode in which the cleaning result is automatically executed by using the judgment result of the chip information judging unit 46 as it is.
- the teaching mode is a mode in which the cleaning process is executed by using the determination result corrected by the teaching processing unit 49.
- the judgment parameters are stored in the judgment parameter storage unit 32 in advance.
- one or a plurality of sample images of chips are prepared, and the sample images are divided into a predetermined mesh size.
- a large number of mesh images are acquired as teaching data for the chip determination model to learn.
- the teaching data is learned by the chip determination model, and the optimized determination parameter is stored in the determination parameter storage unit 32.
- the teaching data used in this embodiment is a mesh image and its class, and a large amount of teaching data is required to obtain a certain level of performance and versatility. For some reason, teacher data is easy to prepare. (1) Many mesh images, which are learning units, can be obtained from one area image. (2) Even with the same area image, different teaching data can be easily increased by performing simple processing such as shifting or rotating the division boundaries of the mesh image. (3) Object detection algorithms such as Object Detection and Semantic Segmentation require information on the position of all chips reflected in the area image. On the other hand, in the present embodiment, by dividing the mesh with a mesh size necessary and sufficient for cleaning, it is not necessary to specify the exact position of the chip in the area image, and the classification can be simplified. (4) When class classification is used as chip information, teacher data is created by assigning one numerical value (class) to each mesh image. ⁇ 0 2020/175 308 18 ⁇ (: 171? 2020 /006748
- the target area setting unit 41 sets the target area in which chips are to be detected (step 3 1), and then the drive control unit 4 2 sets the machine tool. Machining the workpiece by controlling 10 (step 3 2). As a result, as shown in Fig. 1, the chips are scattered on the table 12 and the workpiece, and the image data acquisition unit 43 acquires the area image of the target area from the imaging means 11 (step 33 ).
- step 34 the chip detection processing according to the present invention is executed (step 34).
- the chip detection processing according to step 34 will be described with reference to FIG.
- the mesh size selection unit 44 selects a mesh size (step 3 2 1).
- the mesh size selection unit 44 has a mesh size that allows the shape of chips to be recognized, and that the edge shape of the workpiece or jig that is the background of the chips is simplified. Select one or more candidates that are likely to fit your size. As a result, the chips are easily distinguished from the background, and erroneous detection is suppressed, so that the detection accuracy is improved.
- the mesh division unit 45 divides the area image into a plurality of mesh images with one of the mesh sizes selected by the mesh size selection unit 4 4 (step 3 2 2). Therefore, when the chip information is used, it is only necessary to specify the mesh area corresponding to each mesh image, and it is not necessary to specify the exact position for all chips. For this reason, the amount of calculation for chip detection is greatly reduced.
- the mesh division unit 45 determines whether the boundary images of the partial images overlap or not. Then perform the process of I) or (I). This ensures that the same ⁇ 0 2020/175 308 19 ⁇ (: 171? 2020 /006748
- a plurality of different partial images will not be mixed in the cache image. Therefore, erroneous detection due to difference in lightness between partial images and misalignment is reduced.
- the chip information determining unit 46 determines chip information for one mesh image (step 3 2 3) and repeats until the determination of all mesh images in the target area is completed (step 3 2 4: N 0). As a result, it is only necessary to determine chip information for a mesh image divided into a predetermined mesh size, which simplifies and improves the accuracy of chip detection processing.
- step 3 2 4 Mimi 3
- the chip information determination unit 46 determines whether or not the determination has been completed for all mesh sizes in the target area (step 3 2 5). Then, if there is a mesh size that has not been judged yet (step 3 2 5: 1 ⁇ 100), the process returns to step 3 2 1, and the division process (step 3 2 2) and judgment process (step 3 2 2) with the mesh size are performed. 3 2 3) is repeated.
- step 3 2 5 Mimi 3
- the chip information judgment unit 46 stores the judgment results for each mesh size in the judgment result storage unit 3 3. Then (step 3 2 6), the process is terminated.
- step 35 it is determined whether or not the operation mode is set to the teaching mode (step 35), and if the automatic cleaning mode is set (step 35: N 0) , Proceed to the processing of step 7 described later. On the other hand, if the teaching mode has been set (step 35: step 3), the teaching processing section 49 executes the teaching process (step 36).
- step 36 The teaching process according to step 36 will be described below with reference to FIG.
- the teaching processing unit 49 displays the judgment result on the display input means 2 (step 331) and accepts the selection of the mesh image by the user. At this time, if there are judgment results for a plurality of mesh sizes, the user can visually select the optimum mesh size and use the mesh size to improve the accuracy of chip information judgment. Continue ⁇ 0 2020/175308 20 20 (:171?2020/006748
- step 332 It is determined whether or not a misaligned mesh image is selected (step 332), and if no mesh image is selected (step 332:N0), unless the teaching mode is finished (step 332). 3 5 :N 0), accept selection of mesh image.
- the teaching processing unit 49 causes the user to input the correct chip information about the selected mesh image, and acquires the chip information as a determination result (step 3 3 3).
- the correct judgment result corrected and visually confirmed by the user is stored in the judgment result storage unit 33 (Step 3 34) and the judgment result is displayed on the display input means 2 (Step 3 3 5) 0
- step 3 36 Mimi 3
- the additional learning unit 50 starts additional learning in the background at a predetermined timing, and the corrected determination result is displayed.
- the reflected judgment parameters are acquired and the judgment parameters in the judgment parameter storage unit 32 are updated with the judgment parameters (step 337).
- the cleaning necessity determination unit 47 determines that the chips are cleaned. It is determined whether or not is necessary (step 37). In the present embodiment, since the cleaning condition is set by combining the chip information and the position of the mesh area, the cleaning process is efficiently executed according to the detected amount and position of the chips.
- the cleaning condition may be that only the mesh area corresponding to the mesh image determined to be class 2 (there are many chips) is cleaned. If so, the cleaning condition may be that no cleaning is performed. By setting such cleaning conditions, And the consumption of cleaning liquid is suppressed.
- step S7 As a result of the determination in step S7, if cleaning is not necessary (step S7
- step S13 the cleaning processing unit 48 initializes the number of continuous cleanings and the number of local continuous cleanings. Then, unless the end of machining is instructed (step S 15: N0), the process returns to step S 1 and the subsequent processing is repeated. As a result, it becomes possible to machine the workpiece with the chips removed.
- step S 7 if cleaning is required (step S 8)
- the cleaning processing unit 48 determines whether or not the number of continuous cleanings reaches a predetermined threshold value (step S8). As a result of the determination, when the number of continuous cleanings reaches the threshold value (step S8: YES), the cleaning processing unit 48 executes the abnormal condition processing without performing the cleaning operation (step S14). This avoids the case where the chips that should have been cleaned have just moved to another mesh area but have not actually been removed, and the automatic cleaning operation does not end forever.
- step S 8 even if the number of continuous cleanings is less than the threshold value (step S 8: YES), the cleaning processing unit 48 separately determines whether or not the number of local continuous cleanings reaches a predetermined threshold value. It is determined (step S9). As a result of the determination, even when the number of local continuous cleanings reaches the threshold value (step S9: YES), the cleaning processing unit 48 executes the abnormal condition processing without performing the cleaning operation (step S14). This makes it possible to erroneously detect chips such as those that try to clean chips that are difficult to remove by washing many times because they are sticking to or stuck on the work, etc., jig screws, and dirt inside the machine. However, the case of forever cleaning is avoided.
- step S8:N0, step S9:N0 the cleaning processing unit 48 determines that cleaning is necessary.
- the mesh area corresponding to each of the mesh images is calculated (step S10), and each mesh area is cleaned by the cleaning means 13 ( ⁇ 0 2020/175 308 22 ⁇ (: 171? 2020 /006748
- Step 3 1 chips are cleaned automatically and efficiently according to the chip information.
- the amount of cleaning liquid sprayed is less than when cleaning the entire machine, which is economical and eco-friendly.
- step 33 Return to step 33 and repeat the subsequent processing. As a result, unless the number of continuous cleanings and the number of local continuous cleanings reach the threshold value, the chip information is determined and the cleaning according to the cleaning condition is executed.
- step 3 15: ⁇ 3 when the processing end is instructed (step 3 15: ⁇ 3), this processing ends.
- Chips generated when a workpiece is machined by the machine tool 10 can be detected easily and with high accuracy.
- Chips can be detected with high precision using machine learning.
- Chips can be easily detected using the image feature amount.
- Chips can be cleaned automatically and efficiently according to the chip information. 8 Can be efficiently cleaned according to the amount and position of the detected chips. 9. It is possible to prevent the cleaning operation from being repeated forever due to misdetection of chips.
- the chip detecting device 1 according to the present invention, the machine tool 10 including the same, and the chip detecting method are not limited to the above-described embodiments, but can be appropriately changed.
- the chips are cleaned using the chip information, but the present invention is not limited to this configuration, and the chip detection device 1 has a cleaning function. ⁇ 0 2020/175 308 23 ⁇ (: 171? 2020 /006748
- the chip information may be used in various situations other than being used as information for cleaning the chips. For example, if the mesh information can identify a mesh area where a large amount of chips or the chips are likely to accumulate frequently, it will be used as reference information when setting cleaning conditions such as cleaning the mesh area early or always. Similarly, based on the information on the mesh area where chips are likely to accumulate, it may be used by the machine tool vendor when designing a machine in which chips are less likely to accumulate.
- the chip detection device 1 has the teaching mode, but it is not an essential configuration. If the teaching mode is not required, it is not necessary to provide the teaching processing unit 49, and in the flow chart shown in FIG. 9, the processes of steps 35 and 36 are unnecessary. Also, the teaching process shown in Fig. 11 is not required.
- the cleaning operation is controlled by using both the number of continuous cleanings and the number of local continuous cleanings, but the present invention is not limited to this configuration, and only one of them is used. You may control it. In this case, in the format chart shown in FIG. 9, either step 38 or step 39 is unnecessary.
- the learning parameter is used as the determination parameter, but the present invention is not limited to this configuration, and the chip is determined based on various feature amounts of the mesh image. It may be a threshold for classifying information.
- the above feature amount is not particularly limited as long as it is a feature amount that can represent the features of chips, and the degree of complexity of the edge component in the mesh image, the ratio of the straight line component to the edge component in the mesh image, Various features used in image processing such as brightness of mesh images and brightness statistics can be used. ⁇ 0 2020/175308 24 ⁇ (: 171? 2020 /006748
- a score of the feature amount is calculated for each of the mesh images obtained by dividing the sample image, and the score is aggregated for each class of chip information. It may be set by optimizing or may be set by combining a plurality of feature amounts.
- the "class" indicating the easiness of cleaning of chips is determined as the chip information, but the present invention is not limited to this, and the presence or absence of chips and the presence of chips It may be quantity, chip shape, chip type, or a combination thereof. For example, when the amount and shape of chips are used as chip information, it is possible to set a cleaning condition that even if the mesh area has a small amount of chips, chips with a shape that is difficult to clean are cleaned.
- the chip detection device 1 is configured to be able to additionally learn new teacher data by mouth, but the present invention is not limited to this configuration, and the chip detection device 1 and Teacher data may be transmitted to a learning device connected to the network for additional learning. Further, the teaching data for additional learning from a plurality of chip detection devices 1 may be aggregated in the learning device, and the learning parameter (judgment parameter) obtained by additional learning may be shared to improve versatility. ..
- a learning image for deriving the determination parameter used in the above-described chip detecting device 1 may be generated by the learning image synthesizing device 1200 as shown in FIG.
- the learning image synthesizer 1 2 0 0 includes a chip image storage unit 1 2 0 1 that stores chip images representing various chips, and a background image storage unit 1 2 0 2 that stores background images captured in various aircraft.
- the learning image compositing apparatus 1203 includes an image generating unit 1203 that randomly composes the chip image and the background image to generate a learning image.
- the chip image may be an imaged image of an actual chip without a background. Further, the background image may be a picked-up image obtained by actually picking up the inside of the aircraft.
- the image generator 1 2 0 3 composites the chip images into in-flight photos to create multiple environmental patterns and to create a number of learning patterns, as if they were random, but as truly scattered as possible. ⁇ 0 2020/175 308 25 ⁇ (: 171? 2020 /006748
- the image generation unit 1203 creates scratches or the like on the background image.
- the learning image generated by the learning image synthesizing device 1250 is provided to the learning device 125 and stored in the learning image storage unit 1251. Then, it is used by the determination parameter derivation unit 1 2 5 2 to derive the determination parameter.
- the judgment parameters are provided to the chip judgment model derivation unit 1 2 5 3 and are used to generate the chip judgment model.
- the learning image synthesizer 1200 may provide the chip determination model to the chip detector 1 via the network 1260.
- the learning image synthesizing device 1200 and the learning device 125 0 may be provided in the chip detecting device 1, but in such a case, the chip detecting device 1 can operate at a high speed such as ⁇ II. It is necessary to have resources.
- the chip image storage unit 1201 preferably stores images of various chips having different colors, shapes, and sizes. In addition, it is preferable to store an image of chips with coolant and an image of chips without coolant.
- the background image stored in the background image storage unit 1 0 2 is divided into areas with many chips, areas with few chips, and areas without chips. When combining with the chip image, weighting is applied to each area. You may change and synthesize.
- the image generation unit 123 may add a process such as paint baldness, scratches, and stains to the background image, and then combine the background image with the chip image.
- the size of the chip images may be changed or rotated during composition.
- the image generation unit 1 2 0 3 prepares and pastes correct answer data representing the presence or absence of chips for each area of the mesh image.
- the image compositing device for learning 1 2 0 0 uses various background images of various models, various chip images, and various weights and densities, which are combined to produce various effects (brightness, Applying a coolant mist) creates an enormous amount of teacher data.
- the learning image synthesizing device 1200 may generate learning images by classifying the learning images.
- the chip detecting device 1 is realized as one function of the numerical control device, but the present invention is not limited to this configuration. That is, the chip detection device 1 may be configured by a computer independent of the numerical control device.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mechanical Engineering (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Optics & Photonics (AREA)
- Image Analysis (AREA)
- Machine Tool Sensing Apparatuses (AREA)
- Auxiliary Devices For Machine Tools (AREA)
Abstract
Description
Claims
Priority Applications (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US17/433,873 US12340494B2 (en) | 2019-02-25 | 2020-02-20 | Chip detection apparatus, machine tool, chip detection method, and learning image composition apparatus |
| JP2020542668A JP7634140B2 (ja) | 2019-02-25 | 2020-02-20 | 切屑検出装置 |
| JP2022083536A JP7503593B2 (ja) | 2019-02-25 | 2022-05-23 | 情報処理装置、情報処理方法およびプログラム |
| JP2024093822A JP2024103815A (ja) | 2019-02-25 | 2024-06-10 | 情報処理装置 |
| US19/218,640 US20250285257A1 (en) | 2019-02-25 | 2025-05-27 | Chip detection apparatus, machine tool, chip detection method, and learning image composition apparatus |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2019-031127 | 2019-02-25 | ||
| JP2019031127 | 2019-02-25 |
Related Child Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US17/433,873 A-371-Of-International US12340494B2 (en) | 2019-02-25 | 2020-02-20 | Chip detection apparatus, machine tool, chip detection method, and learning image composition apparatus |
| US19/218,640 Continuation US20250285257A1 (en) | 2019-02-25 | 2025-05-27 | Chip detection apparatus, machine tool, chip detection method, and learning image composition apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020175308A1 true WO2020175308A1 (ja) | 2020-09-03 |
Family
ID=72238859
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2020/006748 Ceased WO2020175308A1 (ja) | 2019-02-25 | 2020-02-20 | 切屑検出装置、工作機械、切屑検出方法、学習用画像合成装置 |
Country Status (3)
| Country | Link |
|---|---|
| US (2) | US12340494B2 (ja) |
| JP (3) | JP7634140B2 (ja) |
| WO (1) | WO2020175308A1 (ja) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210158506A1 (en) * | 2019-11-21 | 2021-05-27 | Fanuc Corporation | Control device and recording medium storing program |
| KR102419738B1 (ko) * | 2021-08-31 | 2022-07-12 | 주식회사 별하늘친구 | 인공지능을 이용하는 얼룩검사 학습을 위한 이미지 데이터를 생성하는 방법 |
| WO2022244070A1 (ja) * | 2021-05-17 | 2022-11-24 | ファナック株式会社 | 稼働状況表示装置、およびコンピュータ読み取り可能な記憶媒体 |
| JP2023168068A (ja) * | 2022-05-13 | 2023-11-24 | Jfeスチール株式会社 | 画像解析装置、機械の評価装置、及び機械の診断方法 |
| CN117400065A (zh) * | 2023-11-08 | 2024-01-16 | 珠海格力智能装备技术研究院有限公司 | 基于图像识别的数控机床的切削质量确定方法及装置 |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7065262B2 (ja) * | 2019-12-09 | 2022-05-11 | Dmg森精機株式会社 | 情報処理装置、工作機械及び情報処理システム |
| JP7762678B2 (ja) * | 2022-03-28 | 2025-10-30 | 日本碍子株式会社 | ワークの検査装置及び方法 |
| JP7782409B2 (ja) * | 2022-10-18 | 2025-12-09 | トヨタ自動車株式会社 | 情報処理装置、情報処理方法、及び情報処理プログラム |
| US12367587B2 (en) * | 2023-01-03 | 2025-07-22 | Fair Isaac Corporation | Segmentation using zero value features in machine learning |
| JP7717108B2 (ja) * | 2023-03-29 | 2025-08-01 | 日本碍子株式会社 | ハニカム構造体の製造方法及び検査装置 |
| KR102892131B1 (ko) * | 2023-12-28 | 2025-11-28 | 주식회사 딥인스펙션 | 생성형 인공지능 기반 인프라시설의 자동 메쉬생성 및 노이즈필터를 포함한 디지털트윈 모델 구축 시스템 |
| WO2025154142A1 (ja) * | 2024-01-15 | 2025-07-24 | ファナック株式会社 | 異常判定装置および工作機械 |
| TWI904704B (zh) * | 2024-06-05 | 2025-11-11 | 國立勤益科技大學 | 機械精密加工之排屑系統節能控制方法 |
| WO2026062989A1 (ja) * | 2024-09-19 | 2026-03-26 | ジヤトコ株式会社 | 発生源特定システム及び発生源特定方法 |
| JP7818689B1 (ja) * | 2024-12-24 | 2026-02-20 | Dmg森精機株式会社 | 工作機械および工作機械の加工室内の洗浄方法 |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03184747A (ja) * | 1989-12-13 | 1991-08-12 | Osaka Kiko Co Ltd | 工作機械における工作物の自動照合認識装置 |
| WO2011036892A1 (ja) * | 2009-09-24 | 2011-03-31 | パナソニック株式会社 | 運転支援表示装置 |
| JP2015026369A (ja) * | 2013-06-20 | 2015-02-05 | 株式会社パスコ | 林相解析装置、林相解析方法及びプログラム |
| JP2016191979A (ja) * | 2015-03-30 | 2016-11-10 | エヌ・ティ・ティ・コムウェア株式会社 | 感性評価装置、感性評価方法、およびプログラム |
| WO2017078072A1 (ja) * | 2015-11-06 | 2017-05-11 | クラリオン株式会社 | 物体検知方法及び物体検知システム |
| JP2017094420A (ja) * | 2015-11-20 | 2017-06-01 | ファナック株式会社 | 工作機械 |
| JP2018024094A (ja) * | 2017-11-14 | 2018-02-15 | ファナック株式会社 | 工作機械の洗浄システム |
| JP2018097506A (ja) * | 2016-12-12 | 2018-06-21 | 株式会社日立製作所 | 衛星画像処理システム及び方法 |
| JP2018156317A (ja) * | 2017-03-16 | 2018-10-04 | 阪神高速技術株式会社 | ひび割れ検出処理装置、およびひび割れ検出処理プログラム |
| JP2019016249A (ja) * | 2017-07-10 | 2019-01-31 | 住友電気工業株式会社 | 判定装置、判定方法および判定プログラム |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3502422B2 (ja) * | 1993-10-06 | 2004-03-02 | 東芝機械株式会社 | 工作機械の切屑除去装置 |
| JP4085844B2 (ja) | 2003-03-03 | 2008-05-14 | 松下電工株式会社 | 画像処理装置 |
| JP4680026B2 (ja) | 2005-10-20 | 2011-05-11 | 株式会社日立ソリューションズ | 画像間変化抽出支援システム及び方法 |
| JP5893719B1 (ja) * | 2014-12-25 | 2016-03-23 | ファナック株式会社 | 異物の有無を確認する視覚センサを備えたワーク着脱手段付加工装置 |
| JP2017107455A (ja) | 2015-12-10 | 2017-06-15 | キヤノン株式会社 | 情報処理装置、制御方法、及びプログラム |
| RU2694021C1 (ru) * | 2015-12-14 | 2019-07-08 | Моушен Метрикс Интернешэнл Корп. | Способ и устройство идентификации частей фрагментированного материала в пределах изображения |
| JP6557183B2 (ja) * | 2016-06-28 | 2019-08-07 | ファナック株式会社 | 切削加工工具の寿命判定装置、寿命判定方法及びプログラム |
| JP6502983B2 (ja) * | 2017-03-15 | 2019-04-17 | ファナック株式会社 | 洗浄工程最適化装置及び機械学習装置 |
| CN109426771A (zh) * | 2017-08-24 | 2019-03-05 | 日立汽车系统株式会社 | 对车辆周围的小物体区域进行识别的装置和方法 |
| US10692221B2 (en) * | 2018-07-13 | 2020-06-23 | Adobe Inc. | Automatic trimap generation and image segmentation |
| JP6922051B1 (ja) * | 2020-08-06 | 2021-08-18 | Dmg森精機株式会社 | 情報処理装置、工作機械及びプログラム |
| JPWO2022044882A1 (ja) * | 2020-08-24 | 2022-03-03 |
-
2020
- 2020-02-20 WO PCT/JP2020/006748 patent/WO2020175308A1/ja not_active Ceased
- 2020-02-20 JP JP2020542668A patent/JP7634140B2/ja active Active
- 2020-02-20 US US17/433,873 patent/US12340494B2/en active Active
-
2022
- 2022-05-23 JP JP2022083536A patent/JP7503593B2/ja active Active
-
2024
- 2024-06-10 JP JP2024093822A patent/JP2024103815A/ja active Pending
-
2025
- 2025-05-27 US US19/218,640 patent/US20250285257A1/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03184747A (ja) * | 1989-12-13 | 1991-08-12 | Osaka Kiko Co Ltd | 工作機械における工作物の自動照合認識装置 |
| WO2011036892A1 (ja) * | 2009-09-24 | 2011-03-31 | パナソニック株式会社 | 運転支援表示装置 |
| JP2015026369A (ja) * | 2013-06-20 | 2015-02-05 | 株式会社パスコ | 林相解析装置、林相解析方法及びプログラム |
| JP2016191979A (ja) * | 2015-03-30 | 2016-11-10 | エヌ・ティ・ティ・コムウェア株式会社 | 感性評価装置、感性評価方法、およびプログラム |
| WO2017078072A1 (ja) * | 2015-11-06 | 2017-05-11 | クラリオン株式会社 | 物体検知方法及び物体検知システム |
| JP2017094420A (ja) * | 2015-11-20 | 2017-06-01 | ファナック株式会社 | 工作機械 |
| JP2018097506A (ja) * | 2016-12-12 | 2018-06-21 | 株式会社日立製作所 | 衛星画像処理システム及び方法 |
| JP2018156317A (ja) * | 2017-03-16 | 2018-10-04 | 阪神高速技術株式会社 | ひび割れ検出処理装置、およびひび割れ検出処理プログラム |
| JP2019016249A (ja) * | 2017-07-10 | 2019-01-31 | 住友電気工業株式会社 | 判定装置、判定方法および判定プログラム |
| JP2018024094A (ja) * | 2017-11-14 | 2018-02-15 | ファナック株式会社 | 工作機械の洗浄システム |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210158506A1 (en) * | 2019-11-21 | 2021-05-27 | Fanuc Corporation | Control device and recording medium storing program |
| US11657490B2 (en) * | 2019-11-21 | 2023-05-23 | Fanuc Corporation | Control device for controlling machine tool capable of alarm stop and non-transitory computer readable medium recording a program |
| WO2022244070A1 (ja) * | 2021-05-17 | 2022-11-24 | ファナック株式会社 | 稼働状況表示装置、およびコンピュータ読み取り可能な記憶媒体 |
| JPWO2022244070A1 (ja) * | 2021-05-17 | 2022-11-24 | ||
| JP7688121B2 (ja) | 2021-05-17 | 2025-06-03 | ファナック株式会社 | 稼働状況表示装置、およびコンピュータ読み取り可能な記憶媒体 |
| KR102419738B1 (ko) * | 2021-08-31 | 2022-07-12 | 주식회사 별하늘친구 | 인공지능을 이용하는 얼룩검사 학습을 위한 이미지 데이터를 생성하는 방법 |
| JP2023168068A (ja) * | 2022-05-13 | 2023-11-24 | Jfeスチール株式会社 | 画像解析装置、機械の評価装置、及び機械の診断方法 |
| JP7582250B2 (ja) | 2022-05-13 | 2024-11-13 | Jfeスチール株式会社 | 画像解析装置、機械の評価装置、及び機械の診断方法 |
| CN117400065A (zh) * | 2023-11-08 | 2024-01-16 | 珠海格力智能装备技术研究院有限公司 | 基于图像识别的数控机床的切削质量确定方法及装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2022116122A (ja) | 2022-08-09 |
| JPWO2020175308A1 (ja) | 2021-03-11 |
| JP7503593B2 (ja) | 2024-06-20 |
| JP2024103815A (ja) | 2024-08-01 |
| JP7634140B2 (ja) | 2025-02-21 |
| US20250285257A1 (en) | 2025-09-11 |
| US20220237767A1 (en) | 2022-07-28 |
| US12340494B2 (en) | 2025-06-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7634140B2 (ja) | 切屑検出装置 | |
| CN1699916B (zh) | 从由机器视觉检查系统执行的检查操作中排除无关特征的系统和方法 | |
| JP6239232B2 (ja) | 高性能エッジフォーカスツール | |
| JP6792980B2 (ja) | マシンビジョン検査システムのインスタンスによって与えられる多露光画像取得を制御する動作を規定するための方法、コンピュータ読み取り可能非一時的記憶媒体及びマシンビジョン検査システム | |
| JP6774588B1 (ja) | 工作機械およびその制御方法 | |
| JP6071452B2 (ja) | マシンビジョンシステムのパートプログラム編集環境内で編集初期化ブロックを利用するシステム及び方法 | |
| CN114746213B (zh) | 显示装置、机床及液体的放出方法 | |
| EP1622082A2 (en) | Method of measuring occluded features for high precision machine vision metrology | |
| EP1653407A2 (en) | Method of filtering an image for high precision machine vision metrology | |
| WO2022044882A1 (ja) | 情報処理装置及び工作機械 | |
| JP2022180272A (ja) | 情報処理装置 | |
| JP2021126713A (ja) | 情報処理装置及び情報処理システム | |
| JP6922051B1 (ja) | 情報処理装置、工作機械及びプログラム | |
| JP6887033B1 (ja) | 画像処理装置、工作機械および画像処理方法 | |
| JP6774587B1 (ja) | 工作機械 | |
| JP6788758B1 (ja) | 情報処理装置および工作機械 | |
| US9167215B2 (en) | Machine vision system editing environment for a part program in which a continuous stream of image acquisition operations are performed during a run mode | |
| CN114746212B (zh) | 显示装置、机床及液体的放出方法 | |
| CN111275022B (zh) | 基于遗忘因子型经验模态分解的污渍检测分析方法及应用 | |
| CN113592800A (zh) | 基于动态扫描参数的图像扫描方法及装置 | |
| Neufeld et al. | Mixed Initiative Interactive Edge Detection. | |
| WO2025022595A1 (ja) | 切粉堆積領域推定装置、切粉堆積領域推定方法及び切粉除去装置 | |
| Dayrit et al. | Vision Algorithm and Optimization Process | |
| JP2023140007A (ja) | 切削加工システム、制御方法、プログラム及び制御システム | |
| CN121776938A (zh) | 一种铝合金框类零件自清洁方法、系统及介质 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| ENP | Entry into the national phase |
Ref document number: 2020542668 Country of ref document: JP Kind code of ref document: A |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20763882 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20763882 Country of ref document: EP Kind code of ref document: A1 |
|
| WWG | Wipo information: grant in national office |
Ref document number: 17433873 Country of ref document: US |