26 results about "Blob analysis" patented technology

The invention relates to the technical field of industrial quality inspection, provides a wafer defect analysis method and electronic equipment, and is used for improving the wafer defect detection efficiency and precision. For a wafer whose appearance defect detection yield is lower than a preset yield threshold, mapping graphs corresponding to detection results of each core particle on the wafer are used as input of a defect segmentation model, independent defect areas of different defect types are identified, blob analysis is carried out on each defect area by using a computer vision technology, defect geometric information is obtained, and the defect geometric information is used as a defect segmentation model. According to the method, the defect geometric information of the wafer is obtained, whether the wafer has process abnormity or not is judged according to the comparison result of the defect geometric information and the defect card control parameters corresponding to the defect types, AI reasoning based on deep learning and traditional CV card control are combined, generalization of AI and accuracy of CV are fully utilized, and the defect detection precision and efficiency are improved.

A computer-implemented, in particular user-guided, method for selecting a photographic image (F) for use in production control, wherein the method comprises the following steps: a) Receiving from a multitude of input images (E1,...,En) specifying an object to be captured (30) in a multitude of lighting settings, wherein the lighting settings are specified by different configurations of an RGB-W lighting unit (20) and include a white image (E0); b) Selecting at least one candidate recording (K1,...,Km) from the initial recordings (E1,...,En) based on a first selection criterion; c) Defining a template image section (A0) within the white image (E0); d) Generating at least one target image section (A1,...,Am) in the at least one candidate image (K1,...,Km) that corresponds to the template image section (A0); e) Applying at least one image processing algorithm, in particular contour recognition, blob analysis, region analysis, code analysis or model matching, with a first set of parameters and / or parameter values ​​to the at least one target image section (A1,...,Am) to generate processed candidate images (K'1,...,K'm); f) Selecting at least one preselection recording (V1,...,Vk) from the set of processed candidate recordings (K'1,...,K'm) based on a second selection criterion; g) Check whether a pre-selection recording (V1,...,Vk) exists that meets a final screening criterion and, if so, select such a pre-selection recording (V1,...,Vk) as the final recording (F); otherwise h) Repeating steps e) to g), wherein the at least one preselection recording (V1,...,Vk) is used as the at least one candidate recording (K1,...,Km) and in step e) a second set of parameters and / or parameter values, in particular specified by a user, is used;wherein at least one of the selection criteria depends on an evaluation result of at least one image analysis algorithm, namely contour recognition, wherein a collection of reference images is used for the evaluation of the at least one selection criterion, wherein contours are extracted from object edges of the reference images during contour recognition and stored as reference contours, which are searched for in the multitude of input images, wherein the method further comprises calculating a value that indicates a match with a reference contour and wherein, if the value exceeds a threshold, the area under consideration is regarded as a match, wherein an image is rated as good if a predefined number of matches are found and wherein the image is selected as at least one candidate image if the image is rated as good.

The invention relates to a binary image rectangular structure element rapid expansion method based on run-length coding, which comprises the following steps of: 1, traversing an image for one round according to a run-length coding mode to carry out run-length coding to obtain a run-length coding area; step 2, translating the run-length coding region to obtain a translated run-length coding region; step 3, performing transverse sliding expansion on each cluster in the translation stroke coding area; 4, calculating the number of iterations of longitudinal expansion and a termination condition, wherein the longitudinal expansion comprises exponential expansion and non-exponential expansion; 5, calculating the result of the exponential expansion; and step 6, calculating non-exponential expansion, and outputting a final expansion result. The invention provides a rectangular structure element rapid expansion operation algorithm based on run-length coding, and the algorithm scheme can effectively improve the efficiency of Blob analysis and defect detection and assist industrial detection.

This invention relates to the field of image processing technology, specifically to a method for detecting through-hole chucks that integrates mask positioning and AI re-inspection. The method involves preprocessing and contour detection of the original workpiece image to achieve coarse positioning of the through-hole region; constructing a multi-layer mask based on the precise geometric information of the through-hole; extracting candidate defect regions through mask subtraction and bitwise AND operations; performing Blob analysis on the candidate defect regions; and conducting multi-dimensional feature analysis and screening based on the physical prior conditions of the chuck to obtain traditional detection judgment results; inputting the through-hole region image into a deep learning model for re-inspection to obtain deep learning detection results; and performing a joint judgment based on the traditional detection judgment results and the deep learning detection results to output the final defect judgment result. This invention achieves precise positioning of the through-hole through multi-layer mask operation and, combined with the joint decision-making of Blob analysis and AI re-inspection, significantly improves the accuracy and robustness of through-hole chuck defect detection while reducing the false detection rate.

This mounting assistance device comprises: a recognition unit that recognizes a component by using blob analysis on image data acquired by imaging a supply region, to which a plurality of components are supplied, of a bulk feeder; a setting unit that sets an inspection region in the image data on the basis of the blob generated in the blob analysis; and a determination unit that determines the quality of the blob or the quality of the recognized component on the basis of a luminance value of the inspection region.

This invention relates to the field of industrial defect detection technology, and proposes a cross-scale industrial surface defect detection method and system, including the following steps: acquiring template image I a And the industrial product image to be inspected I b And preprocess it; based on the Blob algorithm, the preprocessed image I a I b Feature extraction is performed to obtain image features I'. a 、I' b The image I is estimated using a Homography estimation neural network. a and I b and image features I' a and I' b Perform alignment calibration separately; for the aligned and calibrated image I a I b The difference image is processed by subtraction, and a sliding window is used to filter noise from the difference image to obtain the industrial surface defect detection result. This invention introduces a Blob analysis algorithm to locate small-scale defects in multi-scale defects, and combines it with a Homography estimation neural network to align and calibrate the standard sample image used for the defect sample to be detected, so as to achieve better difference results.

The application provides a CELL edge defect detection method and system based on deep learning, and relates to the technical field of liquid crystal display screen defect detection. A plurality of to-be-detected images are imported; an original lightweight segmentation model DDRNet is changed into a revised segmentation model Rep-DDRNet with a single branch structure, feature detection is performed on the to-be-detected images, and suspected crack areas of the to-be-detected images are determined; a plurality of single-branch structure revised segmentation models DDRNet are called in parallel to perform GPU parallel operation on the plurality of to-be-detected images, and a plurality of suspected crack areas are obtained; the plurality of suspected crack areas are classified one by one through an image classification model Rep-VGG, and interference type and real crack type judgment is performed; crack feature extraction is performed on the real crack image through machine vision Blob analysis, and defect grading is performed one by one, and images greater than a set level are alarmed. The deep learning segmentation model is optimized, the F1-score accuracy of the test data set is improved by 6%, and the overall operation time is saved by more than 1 / 3 through GPU operation according to a parallel architecture.

The application discloses a kind of to support back-end automatic sorting of power board card warehousing detection method related to the technical field of power equipment intelligent warehousing and visual detection.The application is realized by material box scanning code identification, type area division, two-dimensional code reading and whitelist filtering, multi-level brightness Blob analysis, slot and width determination, cross-slot continuation determination, total number verification and data binding upload, and realizes the full-process automation detection of power board card warehousing.The application is compatible with different types of material box, supports different slot and width detection, can still accurately determine whether the board card is occupied or not when barcode identification fails, determines the position information of the target board card to be sorted, and sorts the target board card according to the position information.The application solves the pain points of traditional warehousing detection, such as relying on manual, easy to miss detection, unable to adapt to multi-specification board card, and unable to link automatic sorting, significantly improves the efficiency, accuracy and intelligent level of power board card warehousing.

The invention discloses an E buckle detection method and detection device.The method comprises the following steps of detection device providing, stamping test, image source obtaining, BLOB analysis, point-to-point measurement, condition detection, branch simulation and output image detection structure, stamping test can be conducted on a radiator firstly, so that an E buckle can be completely clamped into a stud or completely withdrawn from meshing, and then the radiator can be directly subjected to the stamping test; and the unstable meshing state of semi-buckle connection is avoided, the E buckle assembly condition of each workpiece is automatically judged in combination with a visual detection method and is filed and reserved, and manual detection is avoided.

The invention discloses a metal tube end defect detection method and system based on machine vision, and the method comprises the steps: constructing a metal tube end defect detection device, obtaining a to-be-detected metal tube image, carrying out the image preprocessing, and obtaining a preprocessed to-be-detected metal tube image; an adaptive weight updating mechanism is introduced to optimize a parallel grey wolf optimization algorithm, and a PGWO-AW image segmentation algorithm is constructed; segmenting the preprocessed to-be-detected metal tube image based on a PGWO-AW image segmentation algorithm to obtain a segmented to-be-detected metal tube image; and performing spot analysis on the segmented to-be-detected metal tube image to obtain a metal tube end defect detection result. Defects such as burrs at the end of the metal pipe and scratches on the surface of the end can be automatically detected, and the detection efficiency and accuracy are improved. The metal pipe end defect detection method and system based on machine vision can be widely applied to the technical field of metal pipe detection.

The invention discloses an image corrosion method and device, electronic equipment and a storage medium, and the method comprises the steps: traversing an obtained current image and all runs of a structural element, generating a basic run, and marking the serial number of the run of the structural element corresponding to the basic run, the basic run is a result of executing corrosion operation on the run of the current image and the run of the structural element; the basic runs are traversed line by line, if the number of the basic runs of the current line is smaller than M, or the number of the basic runs of the current line is larger than or equal to M and the structural element run indexes are incomplete, all the basic runs of the current line are deleted, and M is an integer larger than 1; otherwise, in the current row, performing intersection operation on the basic run according to the structural element run index, and searching the public area of the structural element run index. By adopting the technical scheme, the Blob image corrosion method based on run length coding is introduced, so that the Blob analysis processing speed and efficiency are efficiently and remarkably improved.

The invention discloses a steel spline grabbing method and system, and the method comprises the steps: carrying out the color space conversion of a designated region of an original image of a steel spline, and outputting a designated color channel gray image; finding a target position in the specified color channel grayscale image through a target detection algorithm, and carrying out target classification; detecting and segmenting a target classification result by adopting an image segmentation model to obtain a steel spline image; performing BLOB analysis on the steel spline image to obtain steel spline camera coordinates; the steel spline camera coordinates are calibrated and converted to obtain mechanical arm world coordinates of the large spline grabbing robot; and the steel sample strip is grabbed according to the world coordinates of the mechanical arm. According to the invention, a robot automatic sampling assembly line system in a steel production and manufacturing process can realize intelligent grabbing of steel splines, so that the working efficiency is improved, and the labor cost is saved.

The invention relates to the technical field of crystal angle identification, in particular to a crystal angle identification method based on electrode features, which comprises the following steps: generating a multi-modal feature template, matching a crystal target in an original crystal image with the multi-modal feature template, and outputting a matching result; establishing an offset reference; based on an offset reference, dynamically correcting the pixel point position of the region of interest subjected to the Blob analysis after optimization and adjustment, and performing primary Blob analysis on the original crystal image in the region of interest subjected to the Blob analysis after correction to obtain a binary image of the crystal electrode; performing adaptive multi-scale morphological enhancement and feature enhancement on the binary image of the crystal electrode to obtain an enhanced crystal electrode image; and performing secondary Blob analysis on the enhanced crystal electrode image, extracting angle features, and completing identification of the crystal angle based on the electrode features. According to the method provided by the invention, the crystal angle can be accurately and quickly identified.

The present application provides a kind of defect evaluation method of flexographic first piece, including character detection and dirty detection, wherein the character detection includes the following steps: S1: based on the character position of electronic sample and flexographic first piece corresponding construction electronic sample character library and flexographic first piece character library;S2: respectively input the pre-constructed depth evaluation network of electronic sample character and flexographic first piece character, obtain character detection result, complete character detection;The dirty detection includes the following steps: the image of flexographic first piece is subjected to time-frequency conversion, Gaussian filtering and Blob analysis, to obtain dirty detection result, complete dirty detection;According to character detection result and dirty detection result, obtain the defect evaluation result of flexographic first piece.The present application provides a kind of defect evaluation method of flexographic first piece, solves the problem that current flexographic first inspection adopts artificial quality inspection and is easy to produce misjudgment.

The application discloses a kind of element pin detection methods based on machine vision, comprising the following steps: step one: using industrial camera obtains the 3D point cloud of the detected element including pin;Step two: based on the 3D point cloud obtained in step one, obtain height map;Step three: in height map, select a height reference value, and form a selection interval on the height reference value according to the set height difference value;Step four: the point cloud of the sampling point in the selection interval determined in step three is used to form a gray scale image;Step five: in the gray scale image, the center of gravity of the pin is obtained using Blob analysis algorithm and center of gravity algorithm;Step six: use the center of gravity of the pin obtained in step five to judge defects, so use height map to set selection interval to select effective point cloud, so noise point cloud can be filtered, avoid noise point cloud adversely affect gray scale image, to form the real gray scale image of pin, and then judge the defects of pin.

The present invention relates to the technical field of crystal angle recognition, and in particular to a crystal angle recognition method based on electrode features, comprising the following steps: generating a multimodal feature template, matching a crystal target in an original crystal image with the multimodal feature template, and outputting a matching result; establishing an offset reference; dynamically correcting the pixel positions of an optimized and adjusted region of interest of a Blob analysis based on the offset reference, performing a first Blob analysis on the original crystal image within the region of interest of the corrected Blob analysis to obtain a binary image of the crystal electrode; performing adaptive multiscale morphological enhancement and feature intensification on the binary image of the crystal electrode to obtain an enhanced crystal electrode image; performing a second Blob analysis on the enhanced crystal electrode image to extract angle features, and completing the recognition of crystal angles based on electrode features. The method provided by the present invention can accurately and quickly identify crystal angles.

The invention discloses a GPU (Graphics Processing Unit) acceleration-based mask defect detection method and related equipment, and the method comprises the following steps: establishing a hardware link between a camera and a GPU through a PCIE (Peripheral Component Interface Express) bus architecture, and constructing a target memory environment of the camera matched with the GPU; image transmission is carried out on multi-channel image data, collected by the camera, of the to-be-detected mask plate based on the hardware link and the target memory environment, the multi-channel image data is transmitted to the GPU, and image transmission follows an asynchronous transmission mechanism; gPU data processing is carried out on the multi-channel image data through a GPU based on a CUDA flow technology, and a defect binary image is obtained; gPU data processing and image transmission are executed in parallel; carrying out data downloading on the defect binary image based on an asynchronous memory copying mechanism of a CUDA (Compute Unified Device Architecture), and downloading the defect binary image to a CPU (Central Processing Unit) connected with the GPU; data downloading and GPU data processing are executed in parallel; and performing defect feature extraction on the defect binary image through the CPU based on Blob analysis to obtain a defect identification result. Therefore, rapid and efficient mask defect detection can be realized.

The invention discloses a black film and black covered edge distinguishing method and device, equipment and a storage medium, and is applied to the technical field of intelligent processing, and the method comprises the steps: processing a to-be-detected image, obtaining an overall detection image comprising a black film and a black covered edge, carrying out the channel extraction of the image, carrying out the Fourier positive transformation of an S channel of the obtained channel image, and obtaining a channel image; carrying out filtering processing on a low-frequency channel, and then carrying out inverse Fourier transform to obtain a target detection area; carrying out threshold segmentation on the target detection area, and carrying out Blob analysis operation to obtain a black frosted bottom film area; and performing filling processing on the black frosted bottom film area by using the pixel values to obtain a target filling area, and performing edge detection processing by using the target filling area to obtain target boundary information of the black film and the black covered edge. According to the method, image operations such as Fourier transform are performed on the to-be-detected image, and features of the black film and the black covered edge are amplified, so that boundary information of the bottom film and the covered edge film is accurately distinguished.

The present disclosure provides a defect detection method, device, equipment and storage medium, by acquiring a component to be tested point cloud and acquiring a target component template point cloud, wherein the target component template point cloud is a reference template point cloud for judging whether the component to be tested point cloud has an attached defect; the component to be tested point cloud and the target component template point cloud are respectively reconstructed into a component to be tested depth map and a target component template depth map; whether the component to be tested point cloud has an attached defect is determined by performing binaryzation processing and Blob analysis on the component to be tested depth map and the target component template depth map, which not only effectively improves the defect detection rate, but also saves a lot of manpower and resources.

Described herein is a computer-implemented method, including: obtaining, using an image capturing device, at least one digital image, each digital image being obtained at a different angle with respect to a surface of a target coating; performing, using an electronic computer processor in an operative connection with at least one filtering unit, a first image analysis on the obtained at least one image to determine at least one bright region within the at least one image by isolating image foreground data from image background data; performing, using the processor, a blob analysis to determine at least one corrupt area within the at least one bright region; and performing additional steps if at least one corrupt area is found. Also described herein are a respective system and a respective computer-readable medium.

The invention discloses a black film offset detection method, device and equipment and a readable storage medium, and is applied to the technical field of data processing, and the method comprises the steps: carrying out the feature screening of a to-be-detected image, and obtaining a metal frame region; determining a black film initial area according to the metal frame area; performing channel extraction on the initial area of the black film, and determining R, G and B of the image; and performing threshold segmentation and Blob analysis processing on the channel of the initial area of the black film, determining a target image, determining blue film information according to boundary contrast information in the target image, and determining whether the black film deviates or not according to the blue film information. Compared with a current method for detecting whether the black film is deviated or not by adopting a manual visual inspection mode, the method has the advantages that channel extraction is carried out on the initial area of the black film, boundary contrast information of the image is enhanced according to channel information, blue film leakage information is determined more accurately, and therefore the accuracy of black film deviation determination is improved.

The invention discloses a blue film defect identification method, device and equipment and a readable storage medium, which are applied to the technical field of data processing, and the method comprises the steps: carrying out the feature positioning of a to-be-detected image, determining all feature regions of a blue film, carrying out the threshold segmentation and Blob analysis of the feature regions of the blue film, determining a normal feature region of the blue film, and carrying out the recognition of defects of the blue film. Determining an initial defect region according to all the feature regions of the blue film and the normal feature region of the blue film; after edge gradient detection is carried out on an S channel of the initial defect area, threshold segmentation and Blob analysis are carried out, and an area with color change is used as a suspected defect area; detecting the suspected defect area by using a feature detection algorithm to determine the blue film defect; and the feature detection algorithm is an algorithm for detecting by using features of the suspected defect area and determining the blue film defect category. According to the method, the to-be-detected image is analyzed and processed, the suspected defect area of the blue film is extracted, the suspected defect area is detected by using the feature detection algorithm, and the blue film defect is determined.

The application provides an online quality sensing method for multiple types of battery pieces in a flower basket, mainly including the following three steps: S1, three-camera-based image acquisition and splicing, S2, based on Blob analysis and morphological principles, obtaining the multiple types of battery piece regions in the flower basket, and S3, matching the characteristic value parameters of the finished and semi-finished battery piece regions, for the obtained finished and semi-finished battery piece regions, the characteristic values such as the center of gravity, length and width are obtained through region decomposition and other methods, and the characteristic values are screened to remove other interference regions, to obtain accurate finished and semi-finished battery piece regions, and then the threshold range of the related characteristic values set in the algorithm is matched, to judge whether the selected finished and semi-finished battery piece regions meet the set standard, so that the online quality sensing of the multiple types of battery pieces in the flower basket is realized quickly and stably.

The invention provides an automatic threshold segmentation and defect identification method and system for a fillet annular region, and relates to the technical field of image processing. Blob analysis is performed on a to-be-detected capacitor product image, four circular hole feature regions are extracted, and a geometric mean value of center points corresponding to all the circular hole feature regions is taken as a reference point; collecting the center coordinates of each circular bead annular area in the to-be-detected capacitor product image, and calculating the center coordinates of each circular bead annular area and the displacement value of the reference point in sequence so as to intercept a color sub-image; performing RGB channel decomposition on the color sub-image to obtain a peak gray value and two trough gray values nearest to the peak gray value, and intercepting the color sub-image to obtain an insulation paper area; carrying out contour extraction on the insulation paper region to extract a region mask corresponding to the rounded-corner annular region; and calculating the Euclidean distance from each edge point to the fitting straight line in the regional mask, and outputting a defect edge point set as a defect contour. According to the invention, the accuracy of defect detection can be improved.

The invention discloses a 3D visual guidance positioning method and device for bill unstacking and electronic equipment, and relates to the field of data processing. The method comprises the following steps: acquiring a top image of a bill stack; determining position information and height information of the bill stack based on the top image of the bill stack; determining cross section information of the bill stack according to the position information and the height information; based on the cross section information, a Blob analysis method is adopted to analyze the top image, and the cross section contour of the paper money stack is obtained; acquiring a side image of the bill stack; based on the section contour, analyzing the side image to obtain three-dimensional point cloud data of the bill stack; and coarse positioning is conducted on the three-dimensional point cloud data, the moving direction of a mechanical arm of the unstacking device is obtained, fine positioning is conducted on the basis of the moving direction, and coordinate information of movement of the mechanical arm is obtained. By implementing the technical scheme provided by the invention, the bill unstacking accuracy is improved.

The present invention provides a method, device, and related equipment for local automatic image calibration, comprising the following steps: preselecting a calibration plate of preset standard specifications; detecting the radius and center distance of black dots based on the calibration plate; performing ROI tracking based on the radius of the black dots; obtaining a calibration area where calibration points are distributed based on the ROI tracking; extracting the edges of the calibration points using image Blob analysis and a Canany edge detection algorithm in the calibration area and calculating the image center coordinates of each calibration point; extracting the coordinates of nine calibration points in the image coordinate system and in the world coordinate system; calculating an affine transformation matrix from the image coordinate system to the world coordinate system using a nine-point calibration algorithm; and multiplying the result obtained based on the affine transformation matrix by the coordinates of the starting target pixel point to obtain the coordinates of the starting actual target point, the actual coordinates of the end point, and the actual distance value. The present invention has high calibration accuracy, high efficiency, and a wide range of applications.