5 results about "Subpixel edge detection" patented technology

This invention relates to the field of pharmaceutical automation equipment and discloses an intelligent sampling robot suitable for drug production. The robot includes a multi-sensor fusion positioning system, a force control coordination system, and a cleanroom structure design. The multi-sensor fusion positioning system comprises a 3D laser positioning module, a binocular vision module, a sub-pixel edge detection module, and a sensor fusion algorithm. The force control coordination system includes a hybrid RNN-LSTM (Recurrent Short-Term Memory) and CNN (Convolutional Neural Network) network, an airbag gripper, a six-dimensional force sensor, and an underactuated rotation mechanism. The cleanroom structure design includes a mechanical gripper, a nano-sterile coating, and a stainless steel IP69K sealing structure. By employing a fusion algorithm combining extended Kalman filtering and fuzzy evidence theory, efficient fusion of multi-sensor data is achieved. Dynamic threshold adjustment is realized through the hybrid RNN-LSTM and CNN network, reducing force control fluctuations and lowering production costs.

The application discloses an oil saw guide plate flatness and groove wear image measurement method and system, relates to the technical field of machine vision image measurement, and the method obtains guide plate multi-source fusion image data through the cooperative collection of a line laser projection and coaxial light source illumination, calculates the flatness deviation by using three-dimensional point cloud reconstruction and reference plane fitting, adopts sub-pixel edge detection and multi-section statistical measurement to quantize groove wear parameters, fuses the flatness and groove wear indexes to calculate a comprehensive wear index and output a grading maintenance suggestion, and automatically generates a detection report, thereby providing an objective quantitative basis for oil saw guide plate quality detection.

The application discloses a kind of fusion improved YOLO and subpixel extraction's slope deformation visual detection method, belong to image data processing technical field, including constructing slope dataset;Improved YOLO model is constructed;Improved YOLO network is trained with slope dataset, and improved YOLO model is obtained;In real monitoring point, image in video is carried out key point detection with improved YOLO model, and key point prediction pixel level coordinates are output;Subpixel edge detection method of Zernike matrix is used to generate the subpixel coordinates of key point, and cumulative displacement monitoring is carried out based on subpixel coordinates.The application keeps the advantages of YOLOv8 algorithm deployment efficiency and model light weight, significantly improves the precision of highway slope displacement monitoring, breaks through the pixel level limit and pursues the displacement perception ability of submillimeter level, easy to deploy intelligent highway slope displacement real-time monitoring, effectively prevent disasters.

This invention discloses a high-precision online visual inspection system and method for LED carrier tape based on a dual-station setup. The inspection system in this invention features an adjustable dual-station configuration to meet the requirements for inspecting the planar appearance and thickness dimensions of the LED carrier tape. The structural design fully considers compatibility, inspection speed, and the maximum permissible operating speed of the carrier tape. Furthermore, addressing the challenges of miniaturized carrier tapes and high precision requirements, a least-squares sub-pixel edge detection algorithm is proposed. This algorithm can effectively detect various feature dimensional parameters on the LED carrier tape, achieving high-speed, high-precision positioning and recognition of feature points on the carrier tape image. To address the issue of defocus blurring when the carrier tape moves beyond the lens's depth of field, a linear compensation algorithm is proposed. This algorithm provides real-time feedback of inspection data, quickly guiding the camera and enabling rapid on-site debugging in industrial settings, ensuring the system can be truly applied in industrial production environments.

This invention relates to a high-precision weld joint inspection method and system, comprising: acquiring photographs of a circular plate-like part to be inspected; determining the center and diameter of the circular plate-like part; performing image preprocessing and coarse positioning on the photographs to obtain candidate baselines and candidate circles; filtering the candidate baselines and candidate circles to obtain measurement baselines and measurement circles; constructing baseline measuring calipers and circular measuring calipers; employing a sub-pixel edge detection algorithm to extract sub-pixel edge points; fitting straight lines and ellipses and calculating distances to achieve weld joint defect detection. This invention automatically filters and measures the inspection results, eliminating manual intervention and achieving high-precision automatic weld joint measurement.