SensorMV Vision AI Documentation

The Line Detection module is used to detect straight line segments within an image. It allows users to define a Region of Interest (ROI) and adjust detection parameters to accurately capture lines. The module outputs the coordinates of the two endpoints, the center point coordinates, and the angle of the detected line segments.

The Numerical Display module is used to overlay numerical data output from other functional modules (such as brightness detection or spot analysis) directly onto an image. It allows users to customize the text's appearance, position, color, and font to make detection data more intuitive and readable.

The Jog Motion module is used to execute specific motion tasks by controlling axes via preset speed, acceleration, and movement paths. It supports starting, stopping, and saving configurations for single or multiple axes.

The Edge Detection module is used to perform edge detection operations on images. It allows users to import images from other functional modules, define specific Regions of Interest (ROI), configure detection parameters, and analyze the resulting edge coordinates (X and Y positions).

The SensorMV AI Cloud Service Platform is a cloud-based industrial AI vision platform powered by deep learning. It provides an intuitive and efficient online environment for image annotation, model training, and deployment, designed to handle various industrial image labeling requirements.

The QR Code Recognition module automatically detects and recognizes QR code information within images. It processes input images to extract content (such as text or URLs) and displays the results. Currently, it specifically supports the QRCode format.

The distortion correction module calibrates images to eliminate geometric distortion caused by camera lenses. It calculates distortion parameters based on the position and size of dots on a calibration board, improving the accuracy of subsequent visual processing tasks.

The Position Correction module is used to adjust the center point (X and Y) and the rotation angle of a target circular area in an image. It allows users to derive these coordinates and angles from either 'Grayscale Matching' or 'Shape Matching' results to ensure precise target positioning during image processing.

The Mask Module is used to select and cover specific areas of an input image. By defining a Region of Interest (ROI), users can perform local processing or block out parts of an image, which helps improve the precision and reliability of subsequent algorithm modules.

The Brightness Detection module analyzes images to provide statistical brightness information, including maximum brightness, minimum brightness, brightness mean, and standard deviation. Users can either analyze the entire image or define specific Regions of Interest (ROIs) to obtain more accurate results for targeted areas.

The Distortion Correction module corrects geometric lens distortion in images by calculating camera distortion parameters based on the position and size of dots on a calibration board. This process eliminates image deformation caused by the lens, improving the accuracy of subsequent vision processing tasks.

The Ellipse Fitting module is used to fit an ellipse to a set of points in an image (typically obtained via edge detection or blob analysis). It outputs key geometric parameters including the center position, lengths of the major and minor axes, the rotation angle, and the circularity of the fitted ellipse. This is primarily used for shape detection, object localization, and precision measurement.