Implementation of Robust HOG-SVM based Pedestrian Classification

Abstract

Achieving pedestrian protection by means of computer vision is not a new topic in the field of computer vision research; however it is still being pursued with renewed interest because of the huge scope for performance improvement in the existing systems. Generally, the task of pedestrian detection (PD) involves stages such as pre-processing, ROI selection, feature extraction, classification, verification/refinement and tracking. Of all the steps involved in the PD framework, the paper presents the work done towards implementing the feature extraction and classification stages in particular. It is of paramount importance that the extracted features from the image should be robust and distinct enough to help the classifier distinguish between a pedestrian and a non-pedestrian, while a good classification algorithm would go a long way in precisely identifying a pedestrian as well as in simplifying the verification stage of the PD framework. The presented work focuses on the implementation of the Histogram of Oriented Gradients (HOG) features with modified parameters that can represent accurate intrinsic information of the image. Classification is achieved using Support Vector Machine (SVM). However instead of employing a readily available SVM library, the linear SVM implemented uses the Sequential Minimal Optimization (SMO) algorithm. The results observed by this HOG-SVM combination show promise to be the best feature extraction cum classification module for a full-fledged PD system. General TermsComputer Vision, Image processing.

Dimension Reduction using Orthogonal Local Preserving Projection in Big data

“…associated with the smallest Eigen values. (ii) Compute k V as the eigenvector associated with the smallest Eigen values of k J and mathematically shown in equation (9).…”

“…The pedestrian recognition in image is a challenging task due to the different kinds of scenarios and illumination conditions occur on urban scenes. There is a need of an efficient feature descriptor such as HOG [7], Local Binary Pattern [8], Gradient Localization Oriented Histogram (GLOH), Scale-Invariant Feature Transform (SIFT) and so on [9] to solve those kinds of problems. In some past research works researchers developed several techniques such as machine learning, deep learning technique and several conventional approaches on pedestrians' detection and classification [10].…”

Small Object Detection with Multiscale Features

“…For example, the Harr feature combined with Adaboosting classifier [14] is availability for face detection. For pedestrian detection, we use the HOG feature (Histogram of Gradients) combined with support vector machine [15] and the HOG feature combined with DPM (Deformable Part Model) [16,17] is often used in the field of the general object detection. However, if there are many different kinds of detected objects in an image, those classifiers will fail to detect the objects.…”

A Portable Driver Assistance System Headset Using Augmented Reality

“…HOG works along with Support Vector Machines to classify the images captured by the NoIR camera. It is worth mentioning that using Support Vector Machines (SVM) in combination with HOG provides a very robust method of detecting humans [11][12] the duo of which has been leveraged upon in this innovation. The edge coordinates are used to mark obstacles, highlighting them on the OLED screen by forming a box around it.…”