Detection and Recognition of Objects in a Real Time

Object recognition is used to find the distinctive objects and also classify those objects in given images. Recognition task is done in various fields such as image processing, computer vision and also pattern recognition. The procedure of the projected approach is, first the input image is converted into gray scale image. Next the image segmentation is done by using clustering method called K-Means clustering. It gives four different cluster images and also displays the different objects presented in that four cluster images to extract the objects presented in the image. The segmentation gives more effective result by giving enhanced segmentation quality and also less computation time. Based on the intensity and texture the features of the extracted objects are extracted. The extracted features of the objects are classified using Multi-class SVM (Support Vector Machine) classification method by matching the features of the extracted objects against predefined objects which is stored in the system. IndexTerm Object Detection, K-Means Clustering, Object Recognition, Multi class SVM Classifier. ________________________________________________________________________________________________________