191 results about "Gray level image" patented technology

The invention relates to the technical field of image detection, in particular to a PCBA board defect detection method and system based on image processing, and the method comprises the following steps: carrying out the meshing calculation of a gray scale deviation after a gray scale image is subjected to Gaussian filtering denoising, generating change rate data, carrying out the statistics of a frequency number, constructing a histogram, combining with an Otsu algorithm, and generating a candidate mask; extracting pixels based on a mask, calculating a gradient modulus, screening edge candidate points, carrying out gradient direction connection and morphological processing to generate a complete edge structure, expanding a connected domain through a region growing algorithm, aligning the connected domain with a template contour, and outputting defect coordinates. According to the method, the defect identification sensitivity is improved through combination of gray level image gridding processing and dynamic threshold calculation, a candidate mask is generated through grid gray level change rate statistics and an Otsu algorithm to avoid over-segmentation missing detection, and the contour precision is improved through combination of gradient modulus difference screening and morphological closed operation optimization. The region growing algorithm and template dynamic alignment reduce deformation misjudgment, and staged dimension reduction and feature enhancement reduce calculation complexity and solve resource waste.

The invention relates to the technical field of solar irradiance prediction, and discloses a solar irradiance prediction method and system based on data fusion, and the method comprises the steps: obtaining a cloud layer gray image, a wind speed vector, a terrain elevation, a slope inclination angle, a solar azimuth angle and an elevation angle, calculating the movement speed and direction of a cloud layer based on the cloud layer image and the wind speed vector, and obtaining a prediction result. Calculating a shielding path in combination with a terrain elevation and a sun position; generating a terrain shielding influence coefficient matrix; fusing a gradient and shielding data to generate a dynamic effect graph; performing cloud-ground shielding analysis based on the dynamic effect graph and a cloud trajectory to obtain cloud-ground coupling influence distribution; using a random forest algorithm to fuse the coupling distribution and the shielding coefficient to calculate an initial irradiance probability, and forming a preliminary prediction result; and inputting the time sequence prediction model, the cloud trajectory and the shielding coefficient into a trained time sequence prediction network, and outputting an optimization result. According to the method, prediction requirements under complex terrains and rapid weather changes can be met.

The invention relates to the technical field of mode recognition and data recognition, and discloses an intelligent visual detection method for non-standard precision part surface microdefects, which comprises the following steps: acquiring surface gray level image data of a to-be-detected part, physically abandoning low-frequency components through discrete wavelet transform, and reserving high-frequency detail components to construct a frequency domain input tensor; constructing a double-flow reconstruction model containing a space domain coding network and a frequency domain coding network, and minimizing the distribution difference of the same feature between double-domain characterization through potential feature space consistency constraint joint optimization; the method comprises the following steps of: calculating a spatial domain residual image and a frequency domain residual image, combining a texture topological residual image extracted by structural tensor characteristic decomposition, and generating a comprehensive abnormal response image through weighted fusion to judge the defect, and effectively inhibiting macroscopic geometric contour interference through frequency domain decoupling and a topological check mechanism on the premise of not needing a standard geometric template. And sensitive perception and accurate identification of weak texture defects on the surface of the non-standard part are realized.

The invention discloses a projection single attitude calibration method, system and device and a medium, and relates to the technical field of computer vision and projection interaction.The method comprises the steps that a plurality of calibration patterns are generated and projected to a calibration plane, gray level images are collected, corner point detection and sub-pixel refinement processing are carried out, and a projection single attitude calibration result is obtained. The method comprises the following steps: acquiring an angular point corresponding relation between a projector image and a camera image, carrying out iterative solution on a projection matrix of a projector and a projection matrix of a camera, separating an internal parameter matrix and an external parameter matrix, calculating a basic matrix and an essential matrix, and decomposing to obtain a rotation matrix and a translation vector of the camera relative to the projector; camera imaging distortion is estimated, a distortion correction coefficient is obtained, and distortion correction is carried out; and counting a re-projection error generated in the correction process, and adjusting parameters of the projector and the camera until the error meets a preset precision requirement. Projector parameter calculation is completed through projection single posture calibration, posture adjustment is not needed in the calibration process, the automation degree is improved, and meanwhile manual intervention is reduced.

The invention discloses a power equipment defect intelligent identification method, system and device and a medium, and the method comprises the steps: collecting an original image of power equipment, and screening the original image of the power equipment to obtain a channel image; carrying out enhancement processing on the channel image and then calculating a gray scale difference value to obtain a gray scale image; converting the gray level image into a frequency spectrum image by adopting fast Fourier transform, and constructing a Gaussian filtering function to carry out convolution and inverse transformation on the frequency spectrum image to obtain a spatial domain image; and performing adaptive threshold segmentation on the spatial domain image, dividing the image into a defect area and a non-defect area, obtaining a segmented image, and performing morphological processing on the segmented image to obtain an electrical equipment defect identification result. According to the method, the problem of aliasing in traditional spatial domain processing is solved, the defect area is accurately extracted, short-time interference and real defects can be effectively distinguished, and the segmentation accuracy is improved.

The invention discloses a hydraulic system intelligent fault diagnosis method based on a diffusion model data enhancement framework, and the method comprises the following steps: firstly providing a diffusion model-based data enhancement framework, carrying out the fault simulation of a hydraulic system through a simulation model, converting a time domain signal into a gray image, and achieving the data enhancement through a diffusion model; carrying out quantitative evaluation on the quality of the generated image, and constructing an enhanced hydraulic system fault data set; based on the proposed framework, a dual-domain adversarial neural network model fused with an attention mechanism is proposed, the weight of a loss function of two domain discriminators is adjusted so as to realize global and local feature adaptive fusion optimization, adversarial training is realized through a gradient inversion layer, and the adaptive fusion optimization of global and local features is realized. Experimental fault state data serve as a test set to be input into a model for fault diagnosis, and the superiority of the method on indexes such as accuracy is verified through comparison of traditional models such as DANN and CNN. In addition, the method is simple and easy to implement, and is suitable for intelligent fault diagnosis of the hydraulic system under the zero sample condition.

The invention discloses a mold quality detection method and system based on image processing, and relates to the technical field of mold quality detection.The method comprises the steps that a high-precision industrial camera is installed over a mold to obtain a mold surface gray level image, and graying and smoothing are conducted on the obtained image; constructing a distribution model according to the reference distance distribution condition of each pixel point in the smoothed image, calculating a period strength degree, preliminarily screening the pixel points according to the period strength degree, and distinguishing the possibility of texture pixel points and non-texture pixel points; by using the difference between normal texture periodicity and crack randomness, through calculating the periodicity intensity degree, screening texture pixel points and non-texture pixel points, screening suspected texture pixel points, calculating the periodicity similarity, screening out a texture region and then carrying out iterative threshold segmentation, the detection problem caused by mixing of cracks and normal textures is solved; and the detection accuracy is improved.

The invention discloses a motor current fault diagnosis method based on a de-noising diffusion probability model, and belongs to the technical field of mechanical equipment state monitoring and fault diagnosis, and the method comprises the steps: obtaining an original motor current signal sample, carrying out the wavelet transformation, obtaining a time-frequency grayscale image, and obtaining a time-frequency grayscale image; dividing the sample into a training sample used for diffusion model training and a test sample of a fault diagnosis model; constructing a diffusion model DDPM, carrying out training by adopting the training sample, and generating a pseudo sample based on the trained diffusion model; constructing a fault diagnosis model MAF-Cnet, and training the MAF-Cnet based on the training sample and the pseudo sample to obtain the trained MAF-Cnet; and inputting a test sample into the trained MAF-Cnet for diagnosis to obtain a motor current fault diagnosis result. The method solves the core problem that the generalization ability of the diagnosis model is insufficient due to scarcity of motor fault samples, improves the diagnosis accuracy, and is wider in application scene.

The invention relates to the technical field of video recognition and analysis, and discloses a video recognition and analysis method based on physical characteristics. Video frame pixels are mapped to a two-dimensional coordinate system with the upper left corner as an original point, and mirror image expansion and median filtering are carried out on a gray level image; constructing a binary image based on a gray threshold value, and analyzing and extracting a target region by using a four-neighborhood connected domain; using neighborhood search and polar angle sorting to close the tracking contour, and generating equidistant re-sampling points based on Euclidean distance and an interpolation method; the curvature of the re-sampling points is estimated through a three-point difference algorithm, and zero denominator is avoided through numerical protection; performing discrete Fourier transform on the curvature sequence to extract a frequency spectrum, and normalizing an amplitude to form a standardized feature vector; and finally, inter-frame similarity is calculated based on the feature vector, and the most similar frame is automatically retrieved. By processing unified data standards in stages, edge noise is suppressed, sampling uniformity is ensured, and feature stability and cross-frame comparability are improved.

The invention provides a ship propulsion motor fault diagnosis method and device and a readable storage medium, the ship propulsion motor adopts a permanent magnet synchronous motor, and the method comprises the steps of collecting normal operation data of the permanent magnet synchronous motor under different working conditions and abnormal data of turn-to-turn short circuit and excitation loss faults; overlapping and sampling the collected three-phase current signals to form a fixed-length sample, and converting the three-phase current signals of the sample into phase comparison current points in a gray image state to form a sample data set; a ship propulsion motor fault diagnosis neural network model is built, the model comprises an input layer, a feature extraction layer, a full connection layer and an output layer, an improved residual network module is arranged in the feature extraction layer, a channel attention mechanism and a pre-activation residual module are fused, and the feature learning ability of key fault features of the ship propulsion motor is enhanced; and based on the trained ship propulsion motor fault diagnosis neural network model, performing early turn-to-turn short circuit and excitation loss fault diagnosis of the permanent magnet synchronous motor.

The invention relates to the technical field of image processing, in particular to a magnesium ore granularity detection method and system based on image processing. The method comprises the steps that a magnesium ore image is collected and preprocessed to obtain a magnesium ore gray level image; processing the magnesium ore grayscale image by using a watershed algorithm to obtain a segmented image; constructing a two-dimensional directional Gaussian kernel for each cleavage direction, and performing convolution on the segmented image by using the two-dimensional directional Gaussian kernel; performing straight line detection on the segmented image after convolution to obtain a crack candidate line; when the crack candidate line and the segmentation line are associated in position, whether the two segmentation regions adjacent to the segmentation line are to-be-merged regions is evaluated according to gray feature values of the two segmentation regions; calculating the gray difference of the boundary of each to-be-merged region, and merging the two segmented regions when the gray is consistent; and calculating granularity data of the magnesium ore based on the merged segmented images. According to the method, the over-segmentation problem generated when the watershed algorithm is used for processing the magnesium ore image containing the cracks is solved.

The invention discloses a two-stage cascaded distributed optical fiber sound wave sensing vehicle trajectory reconstruction method and system, belongs to the technical field of intelligent traffic perception, and aims to solve the problem that geometric accuracy and topological integrity are difficult to consider under the conditions of low signal-to-noise ratio and complex road conditions in the conventional DAS vehicle trajectory reconstruction technology. The method comprises the following steps: in the first stage, carrying out nonlinear enhancement on original vibration data and converting the original vibration data into a two-dimensional space-time grayscale image, inputting the two-dimensional space-time grayscale image into a deep learning semantic segmentation network to generate a high-fidelity track mask, extracting a skeleton through morphological processing, and constructing an initial candidate topological graph; and in the second stage, multi-hop neighborhood kinematics constraint pruning is performed on the initial topological graph, false connection edges are eliminated, global optimization matching is performed by using a coupling cost function to repair fractures, then isolated points are recalled through local geometric scores, and finally a vehicle trajectory graph with complete topology is output. According to the method, geometric accuracy and topological integrity can be effectively considered in complex scenes such as speed change, lane change and multi-vehicle intersection, and the robustness of all-weather traffic flow monitoring is improved.

The invention relates to the technical field of composite material production and detection, in particular to a broken filament detection method, device and equipment for the end face of a carbon fiber spool, and the method comprises the steps: carrying out the image collection of the end face of the spool, and carrying out the graying processing of the collected image, and obtaining a gray image; carrying out image enhancement on the grayscale image, screening pixel points of the end face of a spool in the enhanced image through a grayscale condition, and obtaining a plurality of region sets by segmenting a connected domain; selecting an area with the largest area in the plurality of area sets, and removing pixel points belonging to the yarn drum to obtain a plurality of sub-areas; and screening the areas of the plurality of sub-regions, and if the area of the sub-region is greater than a set threshold value, judging that broken filaments exist. According to the invention, high-efficiency and accurate detection of broken filaments on the end face of the carbon fiber spool can be realized, and the quality and performance of carbon fibers in the subsequent processing process are ensured.

The invention relates to the field of melt state monitoring, in particular to an injection molding machine nozzle melt state monitoring method based on machine vision, and the method comprises the steps: obtaining an original image of melt at an injection molding machine nozzle, carrying out the preprocessing of the original image, obtaining a gray image and a melt region, and calculating the contour complexity representing the melt form based on the melt region; mapping the melt glue area to the gray level image, and calculating the energy and entropy value of the melt glue area; based on the gray level images of the current frame and the previous frame, the flow velocity fluctuation coefficient of the melt glue area is calculated, and the flow velocity fluctuation coefficient is in positive correlation with the fluctuation of the flow velocity die length of the pixel points; and constructing a feature vector by using the contour complexity, the energy, the entropy value and the flow velocity fluctuation coefficient, and inputting the feature vector into a preset prediction model to obtain a prediction result of the melt state. According to the invention, various types of defects such as caking, bubbles and flow marks can be monitored.

The embodiment of the invention provides a thermosensitive gray scale printing method and device, a storage medium and electronic equipment, and is applied to a thermosensitive printer, the method comprises the steps that to-be-printed Mth row of gray scale data is received, the Mth row of gray scale data is the Mth row of data in a K-order gray scale image, M is a positive integer larger than or equal to 1, and K is a positive integer larger than or equal to 2; the M-th line of gray scale data and first reference information are stored in a first memory, and the first reference information comprises the number of points to be heated of each level of gray scale data in the M-th line of gray scale data, a starting byte of each level of gray scale data in the M-th line of gray scale data and an ending byte of each level of gray scale data in the M-th line of gray scale data; and based on the first reference information, sequentially extracting gray scale data of each scale in the M-th line of gray scale data, and controlling the thermal printing head to execute thermal printing operation on the thermal medium according to the gray scale data of each scale in the M-th line of gray scale data until the M-th line of gray scale data is printed.

The invention relates to the image processing field, and especially relates to an airborne infrared image ship target segmentation method comprising the following steps: carrying out LBP characteristic value extraction to obtain an LBP characteristic image; obtaining a mode filtering smooth LBP feature image; performing adaptive threshold segmentation; filling a hole area; multiplying the image with the filled hole region by the airborne infrared gray level image pixel by pixel to obtain a region-of-interest image; obtaining an edge detection image; extracting all boundaries of the edge detection image to create a mask image; segmenting the interest region image by using an improved firefly algorithm to obtain a highlight target segmentation image; and multiplying the highlight target segmentation image and the mask image pixel by pixel to obtain a ship target segmentation result image. The ship texture difference is accurately captured through the LBP features, the improved firefly optimization algorithm is combined, different sea conditions and noise interference can be dealt with, the error segmentation rate is reduced, multi-threshold solution is combined, and the segmentation requirements of complex images can be met.

The invention relates to the technical field of elastography, in particular to an ultrasonic elastography liver fibrosis evaluation method based on artificial intelligence. The method comprises the following steps: analyzing the gray level and gradient distribution of pixel points in a gray level image of each image, and obtaining the fat deposition degree of the liver of a patient in each image; the fat display degree of the liver of the patient in each parameter combination is obtained, and a high-display parameter combination is screened out; according to the imaging depth of each high display parameter combination and the fat deposition degree of the corresponding shear wave elastic image, obtaining a corresponding correction parameter combination; according to gradient features of edge pixel points in different elastic modulus range areas, image resolution corresponding to the correction parameter combination is obtained; obtaining an optimal parameter combination, and obtaining the liver fibrosis risk degree of the patient according to the elastic modulus distribution in the shear wave elastic image under the optimal parameter combination. According to the method, the optimal parameter combination is obtained to carry out elastography, so that the accuracy of liver fibrosis risk assessment is improved.

The invention relates to the technical field of lithium battery production, and discloses a lithium battery binder production monitoring method and system based on the Internet of Things, and the method comprises the steps: collecting data in real time through multiple types of sensors, carrying out the classification according to the types of the sensors, generating time sequence data based on multi-time window dynamic segmentation, and converting the time sequence data into a grayscale image for feature extraction; a self-encoder is adopted to train an anomaly detection model, real-time anomaly is judged in combination with a dynamic threshold value, a correction proposal is generated by matching historical normal data, and a control point is optimized to generate a smooth correction curve; and displaying abnormal data and a correction scheme through a human-computer interface, dynamically marking a high-difference region, and incrementally optimizing a model parameter and a threshold value based on user feedback. According to the method, the anomaly detection precision can be improved, and the success rate of first anomaly repair can be improved.

The invention relates to a scratch detection method, device and equipment for a transparent film and a medium, and the method comprises the steps: firstly carrying out image multiplication processing and dynamic range mapping on a transparent film gray level image, and obtaining a target gray level image through threshold segmentation; calculating a pixel gradient magnitude based on the image, synthesizing an edge gradient image, and generating a first salient image through Gaussian filtering and gray linear transformation; meanwhile, logarithmic transformation is carried out on the original target image to obtain a second salient image; secondly, respectively calculating gray average values of the two salient images, determining a self-adaptive segmentation threshold by combining a preset threshold, extracting a defect region through double-image threshold segmentation, and obtaining an intersection to obtain an initial scratch region; and finally, screening according to a preset area condition to obtain a final scratch area. According to the method, through the multi-feature fusion and self-adaptive threshold technology, the problems of low contrast, uneven illumination and the like in scratch detection of the transparent film are effectively solved, the detection efficiency and accuracy are remarkably improved, and the method has high engineering application value.

The invention relates to the technical field of image processing, in particular to a flower disease and insect pest monitoring method based on machine vision, and the method comprises the steps: obtaining a candidate pseudo boundary set in a gray level image of a to-be-monitored flower leaf, and obtaining a gray level distribution feature of a pixel point in a local analysis region of each candidate pseudo boundary; obtaining a vein feature index of each candidate pseudo-boundary according to the linear structure feature of each candidate pseudo-boundary and the linear structure feature of each candidate pseudo-boundary; according to the vein feature index and the direction feature of each candidate pseudo boundary, screening in the candidate pseudo boundary set to obtain at least one pseudo boundary; in the process of carrying out region segmentation on the gray level image by using the watershed algorithm, the initial distance of each pixel point after distance transformation is obtained, adaptive distance compensation is carried out on the initial distance of each pixel point according to the position relation between each pixel point and each pseudo boundary, the compensation distance is obtained, complete segmentation of a scab region is realized, and the scab image segmentation efficiency is improved. And a reliable quantitative basis is provided for flower disease and pest monitoring and prevention decision making.

The invention discloses a low-altitude unmanned aerial vehicle autonomous cruise method and system based on cloud edge basic model collaboration, and the method comprises the following steps: S1, collecting an environment RGB image through an airborne monocular camera of an unmanned aerial vehicle, and carrying out the preprocessing of the image, and obtaining a preprocessed gray image; s2, a neural scheduler based on deep reinforcement learning generates a scheduling instruction according to the environment data and the network state reasoned by the navigation model at the previous moment; s3, generating a preliminary flight instruction; s4, generating an optimized flight instruction; s5, generating a structured flight instruction; and S6, the unmanned aerial vehicle executes the preliminary flight instruction or the optimized flight instruction or the structured flight instruction. According to the invention, through dynamic on-demand cooperation and intelligent scheduling of the end-edge-cloud three-level model, resource consumption and delay are substantially reduced, and high-robustness and high-safety autonomous cruise of the unmanned aerial vehicle in a complex open environment is realized.

The invention belongs to the technical field of image processing, and particularly relates to a pole adaptive positioning method and system based on image processing, and the method comprises the steps: obtaining edge pixel points in a pole gray level image and the gradient direction of the edge pixel points; estimating an illumination distortion main direction according to the gradient direction, and dividing edge pixel points into a positive edge set and a negative edge set; according to the forward edge set, calculating weighted votes of edge pixel points to gradient direction intersection points and accumulating the weighted votes to a forward accumulator, clustering accumulation results to obtain an initial forward candidate center, and finally performing iterative updating through corrected voting weights based on geometric consistency to obtain forward center estimation; obtaining negative center estimation according to the negative edge set; and determining a circle center positioning point of the pole hole by combining the positive center estimation and the negative center estimation. The method effectively overcomes the positioning error caused by uneven illumination and metal reflection, and improves the positioning precision and robustness.

The invention discloses a casting riser image recognition method based on machine learning, and the method comprises the steps: carrying out the texture enhancement of a casting gray level image based on cross guide filtering, and forming a texture enhancement image data set; parallel straight line textures of the enhanced image are detected in a mode of combining a directional Gabor filter bank and direction consistency analysis; performing connected region analysis on a potential riser region, and performing triple screening through an area range, an aspect ratio range and a brightness contrast ratio; extracting an area feature, an edge feature, a texture feature and a shape feature of the riser candidate region, and marking the type of the riser candidate region to form a training sample set; a random forest classifier is introduced, the training sample set is input into the random forest classifier for training, a riser classifier is obtained, and the types of classified risers in the casting image are obtained; accurate positioning and classification of the casting risers are achieved, technical support is provided for automation and intelligentization of casting cleaning, and remarkable practical value is achieved.

The invention relates to the technical field of image recognition, in particular to a method and a system for automatically detecting defects of a static arc contact based on image recognition. According to the method, the surface image of the static arc contact is acquired and converted into the gray level image, so that the influence of illumination reflection on the detection precision can be effectively reduced, and the identification of the defect area is more accurate. Based on a screening mode of gradient abrupt change point distribution and abrupt change region boundaries, a defect region can be accurately positioned, isolated noise points are eliminated, and the detection robustness is improved. The defect depth is analyzed by combining the gray average value and the standard deviation, so that the physical damage condition of the contact can be evaluated, the relationship between the defect depth and the impact influence can be analyzed by using the current load record, and the source tracing of the defect formation reason is realized. According to the contact resistance analysis mode of the defect area, the influence of different damage types on the conductivity can be accurately identified through reference resistance comparison of the complete area, and data support is provided for fault classification.

The invention belongs to the technical field of computer vision, and particularly relates to a vision-based furniture surface paint spraying defect detection method, which comprises the following steps of: acquiring a gray level image of a furniture surface, dividing pixel points into three types by a K-means clustering algorithm, calculating the number and the average gray level value of each type of pixel points, comparing the gradient value and the gray level value of the pixel points, and calculating the paint spraying defect of the furniture surface. Determining an initial point and a growth trend, dividing the growth trend into an overall trend and an incongruous growth trend according to the coincidence proportion of the growth trend, calculating a contour value, forming a central window by a central point and points in eight neighborhoods of the central point, calculating an abnormal index and a probability index according to the gray values of the central point and pixel points in the central window, and calculating the abnormal index and the probability index; according to the method, the defect probability density is further determined, the confidence coefficient of the defect probability density is calculated in combination with factors such as the defect probability distribution, the smoothness and the pixel point gray average value, whether the furniture surface has the defect and the grade of the defect are judged according to a threshold value, and the detection accuracy of the paint spraying defects in small and irregular shapes is remarkably improved.

The invention relates to the technical field of image data processing, in particular to a cord fabric production defect detection method based on machine vision, which comprises the following steps: acquiring a gray level image of a cord fabric and dividing the gray level image into image blocks; calculating a texture continuity score by fusing the local texture continuity and the periodic rule of the image block, and screening a defect area according to the texture continuity score; extracting the gray level, gradient and texture disorder degree of the defect area to calculate an adaptive weight, and adjusting a cutting limit value of a local contrast enhancement algorithm by using the adaptive weight; performing local contrast enhancement processing on each defect area by adopting the cutting limit value; and classifying the processed defect areas so as to output a defect detection result. According to the method, weak defects can be effectively highlighted, background noise is suppressed, high-precision defect recognition is completed in combination with a deep learning model, and the accuracy and robustness of detection are improved.

The invention relates to the technical field of computer vision, in particular to a method for quickly identifying apparent defects of fish oil soft capsules. The method comprises the following steps: acquiring a reference region in a fish oil soft capsule image to be identified; obtaining a seed point in the reference area and a corresponding growth threshold value, carrying out area growth, and obtaining all foreign matter reference points in the reference area; acquiring an overlapping coefficient of a non-foreign matter reference point in the reference area, acquiring a new seed point and a corresponding new growth threshold when the overlapping coefficient of the non-foreign matter reference point is greater than a preset overlapping coefficient threshold, performing area growth, and acquiring a new foreign matter reference point; all the foreign matter reference points and the new foreign matter reference points in the reference area form a foreign matter area of the reference area, the foreign matter area of the area where each capsule is located in the gray level image is obtained, and defect recognition is conducted on the fish oil soft capsule image. According to the invention, the complete foreign matter area can be rapidly obtained, and the apparent defect identification of the fish oil soft capsule is more accurate.

The invention provides an online defect detection method and device for a silk-screen product, and relates to the technical field of image processing.The online defect detection method comprises the steps that an online detection image of the silk-screen product is obtained and converted into a gray level image, and then silk-screen edge pixel points are extracted; analyzing the edge pixel points to obtain features reflecting the deinking defect, reconstructing the online detection image according to the features to obtain a feature image, taking the deinking defect condition as a label, and training a convolutional neural network by using a historical feature image; and using the trained convolutional neural network to identify whether the feature image has the deinking defect. According to the method, a sub-pixel-level edge retraction and groove structure which is originally difficult to recognize by a traditional algorithm is converted into learnable image depth features with numerical value mutability, the image is reconstructed on the basis, the convolutional neural network under the feature image is constructed, and the sensitivity and generalization ability of a model to structural slight defects are remarkably improved.

The invention belongs to the technical field of radiation detection, and particularly relates to a radiation dose distribution detection device and method when rays pass through a human body, a liquid conveying device conveys a human tissue density simulation liquid to a set liquid level of a detection chamber, and a ray beam passes through the simulation liquid in the detection chamber; a semiconductor detector array with X-ray and proton beam dual-mode detection capability is emitted, the intensity of ray beams emitted from a detection chamber is converted into corresponding output electric signals, then radiation dose distribution conditions when rays enter a human body at different depths are obtained, and visual display is carried out through a three-dimensional gray level image. Based on the same inventive concept, the invention also provides a medium and a system which store a program of the method, and also provides a method for tracking a focus during precise treatment of the proton beam, and the method realizes X-ray real-time monitoring and linkage proton beam parameter regulation and control by utilizing radiation dose distribution of rays extracted by the device in human tissues. The method has the advantages of accurate irradiation positioning, efficient detection, stability, reliability and the like.

The invention discloses a printing defect detection method and device based on cross-modal alignment, and relates to the technical field of printing defect detection.The method comprises the steps that cross-modal feature alignment of a first printing target image (color image) and a second printing target image (gray level image) is achieved, and geometric transformation parameters of the two images are extracted to complete pixel-level registration; the registered image is converted into a grey-scale image, and channel dimension splicing is carried out to generate a spliced image; and detecting the spliced image by adopting an anomaly detection algorithm to realize the positioning of the printing defect. According to the method, the dependence of a traditional method on a reference image is broken through, model training can be completed only through a defect-free sample, the false detection problem caused by a complex texture background, anti-counterfeiting film interference and personalized features in a small sample scene is effectively solved, the time consumed by single-piece detection is less than 0.3 second, and the efficiency is improved by more than 10 times compared with that of traditional manual detection.