The tangent classifier
نویسندگان
چکیده
Given a classifier, we describe a general method to construct a simple linear classification rule. This rule, called the tangent classifier, is obtained by computing the tangent hyperplane to the separation boundary of the groups (generated by the initial classifier) at a certain point. When applied to a quadratic region, the tangent classifier has a neat closed-form expression. We discuss various examples and the application of this new linear classifier in two situations under which standard rules may fail: When there is a fraction of outliers in the training sample, and, when the dimension of the data is large in comparison with the sample size.
منابع مشابه
Experiments with an Extended Tangent Distance
Invariance is an important aspect in image object recognition. We present results obtained with an extended tangent distance incorporated in a kernel density based Bayesian classifier to compensate for affine image variations. An image distortion model for local variations is introduced and its relationship to tangent distance is considered. The proposed classification algorithms are evaluated ...
متن کاملLearning Discriminant Tangent Models for Handwritten Character Recognition
Transformation invariance is known to be fundamental for excellent performances in pattern recognition. One of the most successful approach is tangent distance, originally proposed for a nearest-neighbor algorithm (Simard, LeCun and Denker, 1993). The resulting classifier, however, has a very high computational complexity and, perhaps more important, lacks discrimination capabilities. We presen...
متن کاملRecognition of handwritten digit with transformed invariant distance - Project Final Report - Due 12 / 14 / 07 Buyoung
This paper presents a method to classify handwritten digit based on tangent vectors, which are the linear derivatives of transformations. The purpose of tangent vector is to find the distance between manifolds; a substitute of the classical Euclidean distance. Using the tangent vector, a satisfying performance was achieved, invariant to transformation. While implementing the classifier, we impr...
متن کاملThe Tangent Kernel Approach to Illumination-Robust Texture Classification
Co-occurrence matrices are proved to be useful tool for the purpose of texture recognition. However, they are sensitive to the change of the illumination conditions. There are standard preprocessing approaches to this problem. However, they are lacking certain qualities. We studied the tangent kernel SVM approach as an alternative way of building illumination-robust texture classifier. Testing ...
متن کاملClassification of Radiographs in the ‘Image Retrieval in Medical Applications’ - System (IRMA)
In this paper we present a new approach to classifying radiographs, which is the first important task of the IRMA system. Given an image, we compute posterior probabilities for each image class, as this information is needed for further IRMA processing. Classification is done by using an extended version of Simard’s tangent distance within a kernel density based classifier. We propose a new dis...
متن کامل