43results about "Optically investigating flaws/contamination" patented technology

The invention discloses a detection device applied to bridge cracks. The detection device comprises a flat plate, a first flaw detection device and a second flaw detection device. The first flaw detection device is arranged at the bottom of the flat plate, a walking device is arranged at the bottom of the flat plate, and a control box is arranged on one side above the top of the flat plate. According to the invention, with the second flaw detection device, the inner side and the outer side of the side wall of the guardrail are subjected to crack detection by rotating the lifting device to change different angles; the angle of the second flaw detection device is adjusted through the angle adjusting device; crack detection can be performed on the bottom of the bridge; the inner side and theouter side of the side wall of the guardrail and the bottom of the bridge are detected through the second flaw detection device, the production and use cost is reduced, the application range of the device is widened; the first flaw detection device can move in the length direction and the width direction of the surface of the bridge, dead-corner-free detection on the surface of the bridge is realized, and the working efficiency of the device is improved.

The invention discloses a linear frequency modulation ultrasonic excitation-based infrared thermal wave nondestructive testing method and system. The method comprises the following steps: S1, adjusting spatial positions of an ultrasonic excitation head and a tested sample, and switching on a switch of a cylinder pressing device to ensure that the ultrasonic excitation head is in close contact with the surface of the tested sample, controlling a signal generator to generate a linear frequency modulation pulse signal through a computer, and driving an ultrasonic generator and power amplifier unit to ensure that the ultrasonic power regularly changes according to the linear frequency modulation pulse signal; S2, starting an ultrasonic generator, and acquiring a surface thermal wave signal of the tested sample; and S3, processing the acquired surface thermal wave signal, extracting time-frequency domain characteristic information of the surface thermal wave signal, and integrating the time-frequency domain characteristic information to obtain a characteristic image for characterizing internal defects of the material. The system comprises an ultrasonic excitation device, the ultrasonic generator and power amplifier unit, the signal generator, a focal plane thermal infrared imager and the computer. The internal defects of the material can be rapidly and accurately detected by the method and the system.

The invention provides a method for detecting a flaw on the surface of a fruit by using image processing in the technical field of image processing. The method comprises the following steps of: (1) obtaining a RGB (red, green and blue) image and an NIR (near infrared reflection) image; (2) removing the backgrounds of the RGB image and the NIR image to respectively obtain an RGB foreground image and an NIR foreground image only containing a fruit image; (3) transforming the RGB foreground image into a YCrCb color space image from an RGB color space image, and marking a new image as a new space image; (4) carrying out specific value computation on a Y component image of the new space image and the NIR foreground image to obtain a specific value image; and (5) extracting a fruit flaw image from the specific value image. According to the method, the flaw on the surface of the fruit can be stably, exactly and quickly detected, and the influence caused by the shape and the size of the fruit can be effectively avoided.

The invention discloses a method for scanning wafer defects, which comprises the steps of: calculating to obtain the most frequent unit pixels of various positions on various wafers of some station on a production line and applying to corresponding positions on a virtual wafer to obtain a virtual perfect wafer, and storing the virtual perfect wafer; comparing the stored virtual perfect wafer with the wafer of some station on the production line; and detecting defects of the wafer layer. Furthermore, the invention discloses a machine for scanning wafer defects, including the method for scanning wafer defects. According to the scanning method, a defect scanning way of comparing the virtual perfect wafer and the wafer is added, thereby enhancing the ability of the scanning machine to catch the defects.

The invention discloses a lithium battery coating defect detection method, comprising the following steps: images of a pole piece of a lithium battery are collected; a target image is obtained by splicing the images of the pole piece, and the battery cycle of the lithium battery pole piece is determined; within the range of a battery cycle, a defect region is obtained by extracting defects from the target image; and the eigenvalue of the defect region is calculated, the defect region is classified and matched according to the eigenvalue, and the type of the defect region is determined. The method has low resource utilization rate and high recognition rate. The defect on the pole piece of a lithium battery can be accurately detected, and the defect region can be accurately located and matched.

The invention discloses a part appearance defect detection method, device and equipment and a readable storage medium. The part appearance defect detection method comprises the steps of collecting a second picture containing a to-be-detected part image; determining the pose of the to-be-detected part image in the second picture; generating a second new image according to the pose of the to-be-detected part image and the second image, the second new image including all or part of the to-be-detected part image, the size of the second new image being a preset size, and the pose of the to-be-detected part image in the second new image being within a preset deviation range; and inputting the second new image into the artificial neural network to detect whether the appearance of the to-be-detected part has defects or not.

Provided are methods and systems for inspecting surfaces of various components, such as evaluating height deviations on these surfaces. A method involves aggregating inspection data from multiple line scanners into a combined data set. This combined data set represents a portion of the surface that is larger than the field of measurement any one of the scanners. Furthermore, each pair of adjacent scanners operate at different periods of time to avoid interference. Because operating periods are offset, surface portions scanned by the pair of adjacent scanners can overlap without interference. This overlap of the scanned portions ensures that the entire surface is analyzed. The position of scanners relative to the inspection surface may be changed in between the scans and, in some embodiments, even during the scan. This approach allows precise scanning of large surfaces.

The invention provides a foreign matter detecting apparatus and method. The foreign matter detecting method comprises the following steps: setting the pixel group of a small area including the pixel according to each pixel of the digital image forming the substrate of detecting object; comparing the brightness value of each pixel and the brightness of the pixel group including the pixel; operating the quantity of the deviation degree of each pixel relative to the brightness value of the pixel group according to each pixel via the comparison as the brightness confirmation value; and determining whether exists foreign matter or not based on the brightness confirmation value. The deviation degree of the brightness of the substrate is quite small. As the foreign matter is the object having no fixed shape and naturally formed, the brightness thereof has the characteristics of entirely high brightness level deviation. The difference of the deviation degree of the brightness of the product and foreign matter can be utilized to identify the foreign matter by operating the brightness confirmation value.

The invention discloses a clutch main cylinder groove surface defect detection method based on deep learning. Deep learning and defect recognition are combined to achieve high-precision and high-efficiency automatic detection. A multi-scale feature fusion pyramid network is adopted to extract defect features more accurately, and autonomous learning is carried out; a single-target FCOS convolutional neural network is adopted as a detection network model, dependence on an anchor box and a feature candidate region is avoided, the network model is simplified, and the network training speed is increased; experimental results show that the effect of the defect detection algorithm based on deep learning is superior to that of a traditional manual detection method and a machine vision detection method, and the time requirement and precision of defect detection of the clutch main cylinder inner groove on an industrial production line are met; defects generated in the production process of the clutch main cylinder inner groove can be rapidly and accurately detected, the production efficiency of a factory is improved, and the practical significance is achieved.

The invention relates to a ceramic defect visual inspection method. An SVMD is a user interface and dragging type interactive mode using an image as the center; a concise and efficient work flow process is provided for a user; the system complexity is simplified; the technical requirements on operators are reduced; a system development defect detection module can cover 90 percent or more defect types in industry; particularly, some difficult detection items are overcome, so that the SVMD has the competitiveness in the same industry. The SVMD is widely applied to the industry of parts relevant to ceramics and the like.

A chuck table holding a frame unit including a workpiece is securely placed in an opening of an annular frame by a tape. A transparent holder having a holding surface holds the workpiece with the tape interposed therebetween. A frame body is erected around and surrounding the holder, the frame body having a plurality of suction holes that are open in an inner circumferential surface of the frame body. The frame body has an inside diameter equal to or smaller than an inside diameter of the annular frame. While an opening of the frame body is being covered by the tape, a suction force is transmitted through the suction holes into the frame body, discharging air from between the tape and the holding surface to bring the tape into intimate contact with the holding surface thereby securing the workpiece of the frame unit to the holding surface.

The invention discloses a method for detecting an edge sealing effect of a board. The method comprises the following steps: acquiring first board image information through a first image sensor, wherein the first board image information comprises a first board surface image and an edge sealing band area image; acquiring second board image information through a second image sensor, wherein the second board image information comprises a second board surface image, the second board surface image comprises a laser projection line projected on the surface of the board, and first board image information and the second board image information are the image information of the same board; and performing board defect early warning on the acquired image information. The invention further discloses anelectronic device and a computer-readable storage medium. According to the method for detecting the edge sealing effect of the board, by matching laser projection with image identification, the purpose of detecting multiple different defects at the same time can be achieved, and the detection efficiency and the detection effect of the edge sealing effect of the board are greatly improved.

The invention discloses a device for detecting the surface imperfection of a curved-surface optical element. The device comprises a lighting system and an imaging system and is characterized in that light sources are arranged in the lighting system, the imaging system comprises an imaging lens, an imaging device and a processor in signal connection with the imaging device, the light sources are arranged outside the imaging lens, and the imaging device at least comprises a coaxial imaging mechanism or a lateral imaging mechanism. The device has the advantages that on the basis of the light scattering principle, the light source evenly irradiates the surface of the curved-surface optical element, the imaging system above the curved-surface optical element detects the imaging of the curved-surface optical element, the curved-surface optical elements of different curvature radiuses can be detected, the difficulty that the surface imperfection of the curved-surface optical element cannot beautomatically detected is solved, and the device is low in cost and simple in structure.

The invention discloses a welding spot appearance detection method based on machine vision, belongs to the technical field of welding spot appearance detection and discloses a welding spot appearancedetection method based on machine vision. The method comprises the following steps, firstly, a to-be-detected workpiece is fixed on a conveyor belt; a technician adjusts the positions of the workpieceand the camera; the high-definition camera is close to a workpiece welding spot; then the high-definition camera exposes the workpiece for multiple times to obtain a plurality of high-definition images at the welding spots and transmits the high-definition images to the computer; image processing, the image analysis module performs comparative analysis on the processed image and the template image; when the difference degree between the workpiece image and the template is relatively large; the image analysis module sends an analysis result to the alarm module, Meanwhile, the image analysis module sends the analysis result to a technician, so that welding spot images of a workpiece can be conveniently collected, and an accurate welding spot appearance analysis result can be obtained by comparing and analyzing the welding spot images of different sizes with corresponding templates.

The invention provides a photomask dirt detection system for a laser radar. The system comprises a transmitting device which is disposed on an optical machine rotor of the laser radar, is configured to transmit a detection light beam with a wavelength different from the wavelength of a light beam, emitted by the laser radar, for a detection target, and is used for detecting dirt on a photomask, areceiving device which is arranged on an optical machine rotor of the laser radar, and is configured to receive echoes generated after the detection light beam emitted by the transmitting device is reflected by dirt on the photomask and convert the echoes into a detection electric signal, and a processing unit which is coupled with the receiving device and is configured to identify the dirt condition of the photomask according to the detection electric signal. Through the embodiment of the invention, optical crosstalk can be effectively avoided when the photomask dirt is detected, meanwhile, 360-degree detection of the photomask can be realized, the detection range of the photomask dirt is enlarged, identification of different kinds of dirt is realized, and the detection efficiency is improved.

The invention belongs to the technical field of cylinder sleeve defect detection, and particularly relates to a cylinder sleeve surface defect detection method, system and device based on deep learning, and the method comprises the steps: collecting the real-time surface image data of a cylinder sleeve product on a production line in a regional manner, wherein the product surface image data acquired in different areas at least comprises images of the end face, the inner wall, the outer wall and the skirt edge of the cylinder; filtering and segmenting the image data to obtain to-be-detected area image data of the end face, the inner wall, the outer wall and the skirt edge of the product image; masking the image data of the to-be-detected area to determine the image data of the target detection area of the cylinder sleeve product; and using the trained and optimized classification network model to classify and identify the image data of the target detection area, and acquiring and removing cylinder sleeve products with defects on the surfaces by a sorting system. The invention is combined with deep learning technology to realize rapid and efficient detection of the surface defects of the cylinder sleeve, improves the efficiency, and has a good application value.

The invention discloses a defect recognition device and system based on an image recognition technology. The defect recognition device comprises a detection system module and a shell; illuminating lamps are fixedly arranged at the four corners of the surface of the front end of the shell in an embedded mode; a CCD camera is fixedly arranged in the middle of the surface of the front end of the shell in an embedded mode, and a power socket and a network socket are sequentially formed in the surface of the right end of the shell from front to back; and a display screen is fixed to the upper end surface of the shell in an embedded mode, a storage battery is fixedly arranged at the bottom in the shell through screws, a driving main board is arranged above the storage battery and fixed to the inner wall of the shell through screws, and a detection system module is fixedly arranged on the upper end surface of the driving main board through welding. The defect recognition device and system based on the image recognition technology have the advantages of being high in image information feature online detection and extraction efficiency and accuracy, high in detection precision, capable of labeling defect positions and convenient to compare and check.

The invention provides a connecting rod appearance size detection system. The appearance size detection system comprises a feeding mechanism, a carrying mechanism, a moving mechanism and a laser detection system. The feeding mechanism is connected with the moving mechanism through the carrying mechanism, and the laser detection system is arranged on the moving mechanism. The invention further provides a detection method of the connecting rod appearance size detection system. Through cooperative use of the moving mechanism and the laser detection system, the size, defect detection and characterrecognition of a connecting rod can be efficiently and accurately detected. A plurality of supporting pieces are inserted into two symmetrical clamping grooves. The number and mounting positions of the supporting pieces can be freely adjusted to adapt to connecting rods of different sizes, and the connecting rods are placed on the supporting pieces. The number of the supporting pieces only needsto stably support the connecting rods, the supporting pieces do not need to be laid in a large-area piece-by-piece mode, and therefore material cost is saved, and detection of a follow-up laser detection system cannot be affected.

The invention provides a display screen defect monitoring system based on image recognition and a detection method thereof, and relates to the technical field of screen defect detection. The display screen defect monitoring system comprises a detection assembly line and a main control computer, and the detection assembly line comprises at least one detection station used for detecting a display screen; the detection station is a darkroom in which a lighting jig for loading a display screen and an image acquisition module are arranged, the main control computer comprises an image processing module, an image comparison module, a good product estimation module and a sample library, and a plurality of illuminating lamps are arranged in the detection station. According to the invention, the camera is used for photographing, the image processing module is used for processing the obtained picture, and the image comparison module is used for comparing the processed picture with a non-defective product pre-estimated picture, so that the defective products in the display screen can be effectively detected, compared with the traditional manual detection, the omission ratio and the false detection ratio are greatly reduced, the detection speed is high, the detection precision is high, and the detection efficiency of the display screen is effectively improved.

The invention discloses a working table for display panel testing. The working table comprises an upper supporting block, a lower supporting block, a left supporting block and a right supporting blockwhich are used for fixing a display panel to be detected. The distance between the upper supporting block and the lower supporting block is adjustable. The distance between the left supporting blockand the right supporting block is adjustable. The upper supporting block, the lower supporting block, the left supporting block and the right supporting block define a light transmitting space. The upper supporting block, the lower supporting block, the left supporting block and the right supporting block are provided with fixing parts used for fixing a polarizing plate to the light transmitting space. The working table for display panel testing is adopted for both manual detection and automatic optical detection. Input of equipment is reduced, and the detection cost is lowered.

The invention discloses an engine crankshaft inspection method, which comprises the following steps of: performing magnetic powder inspection on an engine crankshaft by using a continuous method, and judging that the engine crankshaft without magnetic traces on the surface is qualified; judging the magnetic mark position of the engine crankshaft with the magnetic marks on the surface; if the magnetic trace exists on the parting surface (i) or the parting surface and the non-parting surface (ii), judging the shape of the magnetic trace; if the magnetic traces are fibrous, removing magnetic powder, and then using a residual magnetism method for conducting magnetic powder inspection on the engine crankshaft; and if no magnetic trace is found by using the residual magnetism method, determining that the type of the magnetic trace is a streamline type magnetic trace, and determining that the engine crankshaft is qualified. After the engine crankshaft sorted by the inspection method is loaded and used, the fatigue resistance can be comparable with that of a qualified product judged by an existing method, the reliability is high, and no use risk exists in loading. For the same batch of engine crankshafts, the rejection rate in the prior art is more than 10%, and the rejection rate of the test method is only 0.05%, so that the misjudgment rate is greatly reduced.

The invention relates to an automatic focusing device for a linear scanning automatic optical detection system, and the device comprises an optical detection equipment fixing module which is used for fixing optical detection equipment; the wafer lifting bearing module is used for bearing a wafer and driving the wafer to move, and the wafer lifting bearing module is located below the optical detection equipment; and the distance detection module is arranged in the automatic focusing device and can detect the position of the spacing distance between the wafer and the optical detection equipment, and the distance detection module is connected with the wafer lifting bearing module. According to the automatic focusing device, focusing is achieved by fixing the optical detection equipment and adjusting the position of the wafer through the wafer lifting and bearing module, the wafer can be better shot, the setting time is effectively shortened, the optical detection equipment can complete focusing more quickly, and due to the fact that the overall structure of the wafer is light, the wafer is convenient to adjust. And large inertia cannot be generated in the vertical direction, so that the requirement on the torsion of the brake motor is reduced.

The invention discloses a Kingcable white household appliance operation panel vision and function comprehensive test system which comprises a Kingcable white household appliance operation panel vision and function comprehensive test system body arranged in an industrial computer. The Kingcable white household appliance operation panel vision and function comprehensive test system body comprises a user management module, a test data management and storage module, an industrial camera image acquisition module, an image module, an image vision algorithm module, a product function test module, a PLC motion control module and a test item editing module, and the industrial camera image acquisition module is connected with an industrial camera. According to the invention, the design is reasonable, the product quality is detected by machine vision instead of artificial vision, the offline detection efficiency of the product is improved, the working efficiency is improved by integrating machine vision and function testing, and the practicability and universality are enhanced.

The invention discloses a method for counting microparticles on glass. The method comprises steps as follows: (S1) image display: an image of sample glass is input into a computer by an image samplingprocessing device containing a black and white camera, data images are captured in real time through a capture function in EmguCV software and displayed in a control image box; (S2) image enhancement: each frame of the acquired images is preprocessed, so that particles of the images are more prominent; (S3) particle recognition: white bright points which can be seen directly in the black and white camera are subjected to edge detection, the longest edge of each particle is calculated, and the number of current closed regions is counted; (S4) real-time display: an XY coordinate system is drawnon an interface with X representing time and Y representing the number of particles, and the current number of particles is displayed in real time. The method for counting the microparticles on the glass is high in detection efficiency and good in detection effect.