202results about How to "Avoid missing detection" patented technology

The invention discloses a rice seedling leaf age intelligent identification system based on a depth camera, and belongs to the technical field of agricultural intelligent equipment and machine vision. Comprising a monitoring point gridding path planning module, a mechanical arm cooperative positioning data acquisition module, an RGB-D data preprocessing module, a dynamic foreground seedling accurate extraction module, a leaf age recognition confidence evaluation module, a leaf age data visualization presentation module and an integrated control strategy feedback module. RGB color features and connected domain analysis are combined to optimize a mask, a front single seedling is accurately segmented, background interference is eliminated, the confidence coefficient is dynamically calibrated through a deep learning model and image quality features, texture definition and edge continuity features are combined to generate calibrated confidence coefficient, confidence coefficient evaluation is directly associated with an agricultural decision, and the accuracy of the agricultural decision is improved. The leaf age result is generated only based on reliable data, the personnel rechecking demand is reduced, and the decision-making efficiency is improved.

The invention relates to the technical field of coal mill powder pipes, in particular to a powder leakage defect automatic identification method based on digital twinning and AI video analysis, and the method comprises the steps: obtaining operation parameters and environmental parameters of a coal mill powder pipe, calculating a powder leakage risk characterization coefficient, and judging whether there is a potential risk of powder leakage; image data of the surface and the surrounding area of the powder pipe in the initial starting period are collected through an AI video analysis module, powder leakage characteristic parameters are extracted, a characteristic change curve segment is drawn, similarity comparison is conducted on the characteristic change curve segment and the reference characteristic curve segment, and whether a suspected powder leakage area exists or not is judged; actual operation parameters and model prediction parameters of the powder pipe in a stable operation period are obtained based on a digital twin model, the fluctuation degree of parameter deviation is calculated, and whether the powder leakage defect exists or not is comprehensively judged by combining a suspected powder leakage area judgment result; the early-stage accurate identification of the powder leakage defect is realized, the detection efficiency and precision are improved, and the operation and maintenance cost and the safety risk are reduced.

The invention discloses a wafer defect atlas intelligent identification and classification method and system, and relates to the technical field of semiconductor integrated circuit manufacturing and detection, and the method comprises the steps: obtaining a detection point set, an original image and standard reference attribute data; constructing a space weight matrix based on reference data; calculating a generalized connection cost based on a path weighted cumulative value, constructing an initial graph model, and performing probability sparsification; determining verification points based on the weight matrix and extracting local gradient features of the original image; correcting the connection probability by using a Bayesian inference model, and updating a graph topological structure; and finally, outputting a classification result according to the connected sub-graph features. According to the method, layout physical attributes and image microscopic features are combined, and accurate reconstruction and classification of defect forms under a complex background are realized.

The invention provides an intelligent pipeline inner wall damage identification method based on voiceprint feature extraction, and relates to the technical field of sound wave detection.The method comprises the steps that pipeline acoustic time sequence data, medium flow velocity and pressure data, medium type identification and structure parameter information are collected; performing multi-scale feature extraction on the acoustic signal, constructing a voiceprint feature expression model and a damage evaluation model based on a voiceprint feature learning network, introducing a structural voiceprint stability coefficient, a damage type voiceprint separation coefficient and a damage evolution risk coefficient, and combining a multi-level threshold judgment mechanism to determine the damage type of the acoustic signal. Graded early warning, damage type identification and evolution risk discrimination of the inner wall damage of the pipeline are realized; when an uncontrollable damage evolution risk is identified, a maintenance strategy is automatically triggered, and manual nondestructive testing and repairing are guided; and adaptive optimization is carried out on model parameters to form a closed-loop correction mechanism for pipeline inner wall damage identification and risk assessment, so that the pipeline operation safety and the intelligent level of maintenance decision are improved.

The invention provides an excrement sample microscopic image detection method and system based on morphological prior constraint, and the method comprises the steps: obtaining a global background region and a potential target candidate region through separation, so as to generate a standardized category morphological constraint parameter set; inputting the enhanced candidate region feature map into a preset category constraint double-branch instance segmentation network to generate an initial instance mask set corresponding to each independent visible component; performing closed-loop verification of mask integrity and category consistency on the initial instance mask set, performing iterative optimization correction on a mask region which is verified to be unqualified so as to generate an instance segmentation map, and synchronously performing spatial positioning and attribute marking on each independent mask region in the instance segmentation map; and based on an attribute marking result, performing multi-dimensional feature extraction and subdivision category identification on each independent mask region, and synchronously generating a corresponding excrement sample detection report. According to the invention, the precision, stability and complex scene adaptation capability of excrement microscopic detection can be greatly improved.

The invention relates to the technical field of antenna detection, in particular to a laser cutting antenna line type defective product detection method and system, and the method comprises the steps: obtaining processing batch information, building a coiled material motion coordinate system, and storing target antenna parameters and reference information of a dispensing region; acquiring image data of the antenna, and aligning the image data with the reference information in the coiled material motion coordinate system; extracting an antenna boundary according to the image data, and calculating a geometric deviation based on the antenna boundary and the reference information; obtaining a defect coordinate position according to the geometric deviation, and comparing the dispensing area with the geometric deviation to judge the severity of the defect; mapping the coordinate position to a coiled material motion coordinate system to obtain bad position data, and executing differentiation processing according to the severity; and storing the bad position data and the processing batch information corresponding to the differentiation processing, and generating a detection result. According to the invention, the antenna boundary and geometric deviation are analyzed through the image data and the reference information, defects in the antenna cutting process are identified, and the detection accuracy is improved.

The utility model relates to the technical field of bionic pressure sensors, in particular to a broadband wide piezoresistive sensing structure simulating scorpion threshold response characteristics, which comprises a threshold adjusting structure and a piezoresistive sensing unit. The threshold value adjusting structure comprises a flexible deformation cover and a threshold value sensing structure base, the flexible deformation cover covers the top end of the threshold value sensing structure base, the piezoresistive sensing unit is arranged in the threshold value sensing structure base, and a rectangular through groove is formed in the bottom of the threshold value sensing structure base; a threading hole is formed in the side wall of the threshold sensing structure base; the flexible deformation cover comprises a sheet, and the center of the lower surface of the sheet is integrally connected with an inverted-cone-shaped protruding block. The problems that the inherent contradiction exists between the sensitivity and the bandwidth performance of an existing piezoresistive sensor, in particular, structural instability under large signals is easily caused when the sensitivity is improved through a traditional design, and the response capacity to small signals is greatly reduced by increasing rigid support to expand the measuring range are solved.

The invention relates to the technical field of tobacco plant identification and counting, and discloses an unmanned aerial vehicle tobacco plant identification and counting method and system based on improved deep learning, which combines technologies of image classification, image segmentation, feature point extraction, machine learning, target detection and the like with refined intelligent identification of tobacco. According to the method, high-precision image recognition and analysis are realized by utilizing the efficient feature extraction and calculation capability of the method, point plant counting can be accurately performed, the efficiency and precision of agricultural production are greatly improved, the problem of calculation complexity of a traditional method in large-scale high-resolution image processing is solved, and powerful technical support is provided for precision agriculture. Therefore, the invention provides an unmanned aerial vehicle tobacco plant identification and counting method based on improved deep learning target detection and multi-dimensional post-processing, and high-precision and robust detection and counting of tobacco plants in large-scale tobacco field orthoimages are realized by constructing a three-stage assembly line of image preprocessing, improved target detection and multi-dimensional post-processing.

The invention discloses a function detection system and method of an RS232 interface and a storage medium, the system comprises a detection jig and a detected device, and the detected device comprises the RS232 interface; the detection jig comprises a control module, a communication module and a negative pressure detection module. The first side of the communication module is connected with the control module, and the second side of the communication module is connected with the RS232 interface; the first side of the negative pressure detection module is connected with the RS232 interface, and the second side of the negative pressure detection module is connected with the control module; wherein the communication module is used for detecting the communication connection condition of the detected equipment and the detection jig to obtain a first detection result; the negative pressure detection module is used for detecting the detected equipment to obtain a second detection result when the first detection result comprises successful communication connection; and the control module is used for determining a target detection result according to the second detection result. According to the embodiment of the invention, the reliability of the function detection result of the RS232 interface can be improved.

This application provides a method and apparatus for detecting covert communication behavior, relating to the field of network security technology. On one hand, the method uses a dynamic time warping algorithm to determine the minimum cumulative distance between the packet length-timestamp sequence and the standard response traffic template sequence in the packet length dimension. This quantifies temporal misalignment differences, thus solving the problem of missed detection caused by the lack of temporal analysis capabilities in traditional methods. On the other hand, based on the arrival timestamp of the packet length-timestamp sequence, the method determines the differential entropy and the maximum time span of sudden events. The differential entropy can amplify the minute temporal fluctuations caused by the mechanical packet sending of tunnel traffic, while the maximum time span can identify deliberately lengthened idle periods in tunnel traffic. Combining these two methods can accurately capture gradual covert communication such as slow tunnels. Therefore, this method can improve the accuracy of detecting covert communication behavior.

The invention discloses a Kubernetes-based GPU fault automatic monitoring method and system, and relates to the technical field of computer monitoring, and the method comprises the steps: running a GPU detection program at each node; indexes are collected, wherein the indexes comprise XID and / or SXID error events in the system logs and the driving state and the equipment running state obtained based on the NVML; standardizing the indexes into an index data structure; generating a fault judgment result associated with the node identifier and the GPU equipment identifier according to a preset evaluation rule; and when the data is abnormal, the data is mapped as a Node Condition and / or reported to a Kubernetes Event, and a Prometheus acquisition interface is exposed to be pulled and stored. Through the technical scheme of the invention, GPU fault automatic identification and card level positioning are realized, cluster perceptibility and alarm traceability are enhanced, manual troubleshooting cost is reduced, and cluster stability is improved.

The utility model relates to an optical image screening machine, especially based on 3D camera detection's screening machine, it includes preposition detection platform, carousel and loading conveying belt, preposition detection platform includes 3D camera, support plate, loading push plate, receiving material box and discharge push plate, 3D camera is arranged above detection area, loading push plate moves and is set up before and after, discharge push plate moves and is set up left and right, discharge push plate and loading push plate divide the upper surface of support plate into detection area and non - detection area, and the feeding conveyor belt is set up respectively before and after the two sides of detection area with loading push plate, and the receiving material box is set up respectively left and right sides of detection area with discharge push plate, and the carousel is sequentially equipped with supplementary detection device and discharge device along the circumference, and the loading conveying belt links preposition detection platform and carousel, and the bolt is detected in advance in preposition detection platform, can accurately identify various defects, and the bolt that detects the qualified carries out supplementary detection to the carousel, avoids the missed detection, and the equipment occupies the small area.

The present application relates to the technical field of heavy-duty gas turbine, and particularly relates to a heavy-duty gas turbine high-temperature alloy welding defect intelligent identification and repair method. By acquiring a welding area infrared thermal image sequence to identify an abnormal temperature drop area as a candidate defect area, then performing defect verification and type determination through ultrasonic phased array scanning, calculating a repair heat input curve according to the defect type, executing pulse modulation cladding repair and adjusting parameters according to real-time temperature field and acoustic emission signals, automatic identification and accurate repair of welding defects can be realized, and the repair quality and efficiency are improved.

The application discloses a cable surface sheath image preprocessing method, device and equipment and a medium, and relates to the technical field of image processing. The method comprises the following steps: collecting a cable surface sheath image and dividing the image into multiple regions; calculating the average gray value and standard deviation of all pixels in each region, and determining a deviation range based on the 3σ criterion; determining pixels with a gray deviation not within the deviation range as abnormal pixels; determining whether there are other abnormal pixels in the eight-neighbor domain of each abnormal pixel, filtering the abnormal pixel if there are no other abnormal pixels, and constructing a connected domain composed of abnormal pixels if there are other abnormal pixels; filtering connected domains with an area less than a threshold value, and retaining connected domains with an area greater than or equal to the threshold value, to obtain a preprocessed image. Through local adaptive statistics, connected domain analysis and selective processing based on the area, the application overcomes uneven illumination and reflection interference caused by a cable curved surface, and helps to reduce the false positive rate and the missed detection rate of subsequent defect detection.

The application discloses a quality detection system and method for glue injection laser profile, and belongs to the technical field of quality detection. The system comprises the following steps: obtaining a reflected point cloud signal of a glue injection finished product, constructing a point cloud data stream, setting an image reconstruction method, constructing a glue layer three-dimensional model, synchronously generating a pseudo-color gray scale image, and realizing visual representation of the surface texture of the glue layer; constructing a three-layer feature extractor, setting a quality preliminary judgment method, cross-layer fusing features of different scales, solving the missed detection problem of traditional single feature detection on complex defects, automatically identifying basic defects, and outputting a quality preliminary judgment result; judging defects of the glue injection finished product that does not meet the preliminary judgment, setting a defect detailed judgment method, voxel-level segmenting a three-dimensional image, calculating features of the defects, and judging whether the defects affect the use performance of the product, so as to generate a detailed judgment result, realize accurate positioning of complex defects, and improve the defect type recognition accuracy.

The present application relates to a kind of intelligent detection method and detection system of surface coating micron-sized pore, scheme as follows: obtain coating surface image to be detected and execute standardization preprocessing;Preprocessing image is input into trained hole profile segmentation model, and hole profile coordinates are obtained;Based on hole profile coordinates, construct hole connected region by morphological operation, and according to geometric feature, Gap region that does not need to be punched is screened;Preprocessing image is input into trained missed hole detection model, and real missed hole region is obtained;According to hole profile coordinates, calculate geometric feature parameter, determine collapse hole and calculate hole distribution density, finally output detection quality result.The present application realizes the high-precision, automation detection of micron-sized pore by the combination of deep learning model and morphological operation, geometric parameter calculation, solves the inherent defects of artificial detection, meets the high consistency, large batch fast control demand of modern manufacturing industry to surface coating quality.

The application relates to the technical field of pipeline detection, and discloses a pipeline detection device suitable for multifunctional tests, which comprises a bottom plate, an inner supporting circular table I arranged above the bottom plate, an inner supporting circular table II arranged directly above the inner supporting circular table I, a solid interior of the inner supporting circular table I, a hollow interior of the inner supporting circular table II, an open cylinder slidingly connected to the outer wall of the inner supporting circular table I, a hydraulic rod fixedly installed on the inner wall of the open cylinder, a rotating connection between the output end of the hydraulic rod and the bottom of the inner supporting circular table I, a water injection assembly capable of injecting water into the pipeline interior arranged outside the inner supporting circular table II, water injection of the water injection assembly into the pipeline interior through the inner supporting circular table II, full water in the pipeline to simulate the conveying medium state, a pressure medium provided for leakage detection, and a pipeline wall leakage point capable of being detected, the pipeline being in a full water state after water injection, a small amount of water penetrating outward and flowing downward along the pipeline outer wall to provide a detection signal for a humidity detection head if there is a small leakage point on the pipeline wall.

The application provides a kind of based on edge computing and big data's wisdom campus safety early warning method and system. Through edge computing node, real-time collection is carried out to multiple video data streams, frame sequence segmentation and space-time feature extraction are carried out, initial behavior feature set is generated, initial behavior feature set is carried out multidimensional correlation analysis based on preset behavior semantic label, space-time behavior feature vector corresponding to target monitoring scene is extracted, and is transmitted to central server, space-time behavior feature vector is input into target behavior identification model, behavior semantic description sequence corresponding to video data stream is generated, and according to its matching result with preset abnormal behavior rule library, real-time behavior monitoring result is determined, real-time behavior monitoring result is received by edge computing node, and based on dynamic priority strategy, the collection parameters of video data stream are adaptively adjusted.The application can improve the real-time performance, accuracy and system sustainability of campus behavior monitoring.

The invention relates to the technical field of automobile production, and provides an automobile window frame decoration strip detection tool which comprises a frame body, the bottom end of the inner side of the frame body is fixedly connected with a placement seat, a positioning mechanism is arranged above the placement seat, the top end of the inner side of the frame body is provided with a detection mechanism, and the detection mechanism comprises two parallel guide rails. A movable seat is slidably connected between the two guide rails, a sliding block is slidably connected to the inner side of the movable seat, a hollow guide pipe is fixedly connected to the bottom of the sliding block, a detection piece is fixedly connected to the bottom of the hollow guide pipe, a guide plate is arranged on the inner side of the frame body, a guide groove matched with the decoration strip is formed in the guide plate, and a spring is fixedly connected to one end of the inner side of the movable seat. One end of the spring is fixedly connected with the sliding block, and one end of the top of the frame is provided with a translation mechanism. Through dynamic contour tracking, the problem of detection of the special-shaped decoration strip is thoroughly solved, technical breakthroughs are achieved in adaptability, precision and efficiency, and the automobile production quality control cost is remarkably reduced.

This invention relates to the field of image analysis technology, specifically to a machine vision-based quality inspection system for new energy vehicle wiring harnesses and cable ties assembly. The system includes: an image acquisition module that uses polarimetric stereo technology to acquire multi-angle images; a geometric reconstruction and reference extraction module that calculates the surface normal field of the insulation layer and separates the background field of the cable's cylindrical curvature and the micro-indentation residual field, thereby extracting the global axial and radial curvature moduli of the cable; a feature decoupling analysis module that calculates the contour-shaping migration flow field of the indentation micro-residual field relative to the standard contour, and performs anisotropic decomposition of the flow field based on the global axial direction of the cable to obtain the axial slip index; and a quality inspection module that determines an adaptive slip judgment threshold based on the cable's radial curvature modulus, thereby generating inspection results. This invention solves the problems of difficulty in sensing micro-deformation on reflective cylindrical surfaces and the difficulty in decoupling slip risk from assembly tolerances, thus improving the accuracy of wiring harness and cable tie assembly quality inspection.

The present application relates to the technical field of insect screening and collecting, in particular to a virgin fly collecting device and method for fruit fly genetic research; the device comprises a guide pipe and a camera assembly and the like structure, two elastic bags are arranged on the guide pipe, and a preset interval is arranged between the two elastic bags as a detection section; the camera assembly is arranged opposite to the detection section, and is used for shooting photos of fruit flies located in the detection section; real-time image acquisition of fruit flies passing through the detection section is realized through the camera assembly; virgin flies and non-virgin flies are automatically distinguished by combining a deep learning visual recognition program built in an upper computer; and the airflow direction is controlled by an electromagnetic three-way valve to drive the elastic bag and the air blowing pipe to complete the shunting of fruit flies; the whole process does not need manual anesthesia, magnifying glass identification and the like operation, the labor cost is significantly reduced, and the problems of long screening time and large labor intensity of virgin flies in the prior art are solved.

The utility model relates to the technical field of filter leak detection, in particular to a filter leak detection device, which comprises a detection assembly and a track assembly, the detection assembly comprises a scanning head, a measurer, a footstand and a telescopic rod, the scanning head and the footstand are respectively arranged on two sides of the telescopic rod, the measurer is arranged on the telescopic rod and communicated with the scanning head, and the track assembly is arranged on the track assembly. The detection assembly is arranged in the track assembly, the diameter of the opening of the scanning head is set to be larger than the center distance of the convex edges, when the first pulley moves between the convex edges for scanning, the scanning head is arranged in the track assembly, the detection assembly is arranged in the track assembly, and the top end of the scanning head is arranged in an opening mode. And the track assembly comprises a tray, and a plurality of rows of convex edges are evenly distributed on the top face of the tray at fixed intervals. The scanning areas of the adjacent columns / rows are overlapped, missing detection is prevented, meanwhile, the moving distance can be controlled by controlling the distance between the protruding edges, and manpower is saved.

The application relates to a text detection method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: performing a character shape restoration process on a text to be detected to obtain a restored text; performing keyword matching on the restored text based on a keyword library; when the restored text matches a keyword group in the keyword library, obtaining a detection result of the text to be detected according to a keyword segmentation result of the restored text and the matched keyword group; when the restored text does not match a keyword group in the keyword library, performing keyword matching on the pinyin of the restored text based on a pinyin form keyword library, the pinyin form keyword library being obtained by converting keywords in the keyword library into pinyin; and when the pinyin of the restored text matches a pinyin form keyword group in the pinyin form keyword library, obtaining the detection result of the text to be detected according to a pinyin matching restriction condition, thereby improving the text detection accuracy.

The invention discloses a necking detection method based on geometric feature self-adaption. The necking detection method comprises the following steps that S1, the longest main axis of a simulation graph is determined; s2, selecting N feature points on the longest main shaft; s3, obtaining the minimum key size of the simulation graph at each feature point; and S4, judging whether the simulation graph has a necking risk or not based on each minimum key size. By identifying the geometric centroid and the longest principal axis of a graph, adaptively selecting feature points and performing omnidirectional scanning, the measured minimum critical dimension is compared with a set safety threshold, and a risk point which is most likely to generate necking on the graph is accurately positioned, so that the automation and objectification of risk detection are realized, and the risk detection efficiency is improved. Subjectivity and experience dependence of manual marking are avoided, and consistency of detection results is guaranteed; meanwhile, the method has good adaptability to complex or deformed contours, missing detection and misjudgment are effectively avoided, and the yield and reliability of chip manufacturing are improved.

The application belongs to the field of fastener quality inspection, and particularly relates to a sleeve air tightness detection device, which comprises a rack, a linear reciprocating mechanism and an air tightness detection mechanism; wherein the rack is provided with a positioning base for placing one end of a workpiece to be detected with a plug; the execution end of the linear reciprocating mechanism is provided with a gas supply pressure head for plugging the end of the workpiece to be detected away from the plug; the air tightness detection mechanism comprises a controller, a pressure controller connected to a corresponding pipeline, a pressure switch and an electromagnetic valve, and the pressure controller, the pressure switch and the electromagnetic valve are in communication connection with the controller. The sleeve air tightness detection device can replace the traditional manual detection mode, significantly improve the detection quantity per unit time, adopt a standardized detection process and a high-precision sensor, avoid misjudgment or missed detection caused by inconsistent human judgment, reliably detect the sleeve with air leakage defects, and prevent defective products from flowing into subsequent processes or causing faults in field use.

The invention relates to a DPU-based fog cloud collaborative Internet of Things intrusion detection system and method. The system comprises an edge node used for receiving target data sent by Internet of Things equipment, encoding the target data through a pre-trained stacked auto-encoder to obtain a plurality of encoding features, determining a prediction capability index of each encoding feature, and determining a target encoding feature according to each prediction capability index; the cloud server is used for receiving the target coding features sent by the edge node, inputting the target coding features into a pre-trained Internet of Things intrusion identification model, extracting target global features, target local features and target time features associated with the target coding features and Internet of Things intrusion through different branch networks in the Internet of Things intrusion identification model, and sending the extracted target global features, target local features and target time features to the edge node; and outputting an Internet of Things intrusion detection result according to the target global feature, the target local feature and the target time feature. By adopting the system, Internet of Things intrusion detection with low resource occupation, high precision and high efficiency can be realized.

The present application relates to the technical field of computer vision and road engineering, and particularly relates to a pavement crack lightweight segmentation and quantification method based on enhanced feature fusion. The method comprises the following steps: decoding and reading pavement images and performing verification, then performing grid division and cutting, and performing pixel normalization processing on the effective sub-images after cutting; a lightweight crack segmentation network is constructed; the preprocessed standard images are input into the lightweight crack segmentation network, and feature extraction, feature fusion, target detection and pixel segmentation are sequentially completed, and a crack detection frame and a binary segmentation mask are output; the lightweight crack segmentation network is trained, the optimal model is saved according to the iteration optimization of the verification set index; the segmentation mask output by the model is denoised, the crack geometric size is calculated, and the final detection quantification result is output. The method has the advantages that the small-scale crack recognition capability is improved, the crack length and width can be automatically and accurately measured, the crack size measurement error is small, and the method is suitable for large-scale pavement inspection operation.

The invention discloses a power distribution lightning arrester aging defect identification method and device considering environment confidence, electronic equipment and a storage medium, and belongs to the field of intelligent monitoring of power equipment, and the method comprises the steps: collecting the inspection image data of a power distribution lightning arrester and the time sequence synchronous environment parameters of the power distribution lightning arrester; mapping the environment parameters to defect display conditions, quantifying the adaptation degree of the environment to aging defect display, and generating an environment confidence index; visual features are extracted from the inspection image data, similarity calculation is carried out on the visual features and standard defect features, and an image similarity coefficient is obtained; and weighting the image similarity coefficient based on the environment confidence index to form a fusion defect probability value, and outputting a power distribution lightning arrester aging defect identification result according to the fusion defect probability value. Therefore, by implementing the method and the device, the problems of low identification confidence and missed detection and misjudgment caused by incapability of distinguishing'equipment defect-free 'and'environmental implicit' due to lack of quantitative evaluation on environmental suitability in the prior art can be solved.

The invention provides a color feature extraction and correction method and system for drug sensitive plate MIC value interpretation and a computer readable storage medium. The method comprises the following steps: acquiring a to-be-processed drug sensitive plate image; performing visual preprocessing such as exposure, contrast and white balance on the image according to the configuration file; high-precision alignment correction of the image is realized by using ORB feature detection and a BFMatcher matching algorithm; generating a transparent mask with an Alpha channel based on predefined hole position contour data, and performing non-destructive target segmentation on the aligned image; and converting the target region to an HSV color space, extracting H-channel hue features, executing cyclic correction, eliminating the problem of numerical value jump of the red region, and obtaining stable color feature data. The method has the advantages of being high in image registration precision, accurate in segmentation, stable in color feature, high in robustness and the like, and automatic and standardized interpretation of the MIC value of the drug sensitive plate can be achieved under different shooting conditions.

The invention relates to the technical field of vision detection, in particular to an eye position image analysis system for measuring a strabismus angle of a child. The strabismus measurement reliability is determined based on the difference of position changes of different types of pupil center areas in eye position images of two steering eye positions in a symmetrical eye position pair and the difference of the position relations between the pupil center areas and cornea light reflecting spots in the eye position images; the strabismus measurement reliability is adjusted through the included angle between the steering eye position and the orthogonal direction in all symmetrical eye position centering, the strabismus deviation angle is determined, and then the overall measurement reliability is determined; determining a strabismus measurement value according to the difference of position changes of different types of pupil center areas in the eye position images of the steering eye position and the positive eye position; and adjusting the strabismus measurement value by using the overall measurement reliability of all the steering eye positions, and determining the final strabismus angle under the positive eye position. According to the method, the final strabismus angle is determined by integrating the information of all the steering eye positions, and the dynamic characteristics of intermittent strabismus are effectively captured.