Deep Learning VisionPro
A revolutionary step in complex inspection, part location, classification, and OCR
The VisionPro ViDi deep learning algorithms have been developed to analyze real world industrial images. They require a much smaller set of images and much shorter training and validation periods.
The VisionPro ViDi software combines the flexibility of human visual inspection with the speed, repeatability, and power of a computerized system, all in a single, easy-to-use interface.
It has 4 tools:
ViDi Blue-Locate | It locates features
It finds complex features and objects from annotated images.
ViDi Red-Analyze | It detects anomalies and aesthetic defects
It segments defects or other regions of interest simply by learning the varying appearance of the targeted zone. ViDi Red-Analyze identifies scratches on complex surfaces, incomplete or improper assemblies, and even weaving problems on textiles simply by learning the normal appearance of an object including its significant but tolerable variations.
ViDi Green-Classify | It classifies objects and scenes
ViDi Green-Classify separates different classes from a collection of labelled images. Prepared for acceptable tolerances, ViDi Green-Classify identifies products according to their packaging, classifies the quality of welding seams and separates acceptable or unacceptable anomalies.
ViDi Blue-Read | It reads texts and characters
ViDi Blue-Read deciphers incorrectly formed, distorted, or badly recorded codes using Optical Character Recognition (OCR). The previously trained font library identifies most texts without programming or identifying additional texts for fast and easy implementation. This solid tool can be trained to adapt to the specific requirements of the OCR application, it does not require any experience in vision.
Documentation
Technical documentation
