Object Detection and Recognition

Home Services Technologies Object Detection & Recognition
Object recognition in computer vision is a two stage process which involves the task of detecting and classifying a given object in an image or a video sequence. First the objects are localized in the input. Then these objects are described by a set of interest points (feature descriptors). The classification requires building a knowledge base of all classes against which an input object is matched. KritiKal has worked on various systems which are based on the technology of object detection and recognition. Road sign detection, Automatic Number Plate Reading system, Aircraft detection in aerial images, Face recognition, bomb detection in vehicle underside images, Vehicle model detection are few such projects.

 

Features

  • The feature descriptor for the object is invariant to scale, orientation, affine distortion and partially invariant to illumination changes too making the detection and recognition process generic and robust to real world scenarios.
  • Objects to be detected and recognized can vary from a simple shape like a straight line, square, circle, etc to more complex objects like a number plate, face, aircraft, vehicle, building, etc.
  • Handles partial occlusions
  • Intelligent outlier rejection based on various heuristics

 

Applications

  • Vehicle Number plate detection
  • Automatic Vehicle underside monitoring
  • Face Recognition
  • Road Sign detection
  • Pedestrian detection