Orientation Invariant Object Recognitions Using Geometric Moments Invariants and Color Histograms

Abstract

Abstract:Object recognition is a very challenging task in artificial intelligence and robotics. Many approaches have been implemented to achieve this task with greater precision and accuracy. In this paper we have implemented the approach of detecting objects in images undergo with the change in scale, rotation, and orientation. Extracting Geometric moments invariant which are extensively use to extort global features and using color histogram approach we have improved the previously recognition rate to a significant measure. The accuracy of classification is increased by adding the new feature of color Histogram which is also an invariant feature for change in scale rotation, translation, and orientation of objects and using support vector machine learning algorithm for classification.

Determination of constitutive parameters from a Taylor test using inverse analysis

“…The concept of line moments, presented by Lambert and Gao, is an extension of the geometrical moments proposed by Hu to represent the contour data of a scene by a reduced set of parameters. For example, this approach has been used in areas such as pattern recognition, face recognition, ship and aircraft identification, firearm detection, image analysis applications, and in general for object recognition …”

Spatial frequency based video stream analysis for object classification and recognition in clouds

“…Various approaches have been proposed in the past to tackle these problems. The most prominent among these approaches are built on top of insensitive moments [2], color constancy [3] and Fourier phase. However, these approaches are designed to perform classification globally and do not take into account local properties of objects.…”