Convolutional Neural Network Using Kalman Filter for Human Detection and Tracking on RGB-D Video


Jovin Angelico
Ken Ratri Retno Wardani
2018  •  DOI: 10.21512/commit.v12i2.4890

Abstract

The computer ability to detect human being by computer vision is still being improved both in accuracy or computation time. In low-lighting condition, the detection accuracy is usually low. This research uses additional information, besides RGB channels, namely a depth map that shows objects' distance relative to the camera. This research integrates Cascade Classifier (CC) to localize the potential object, the Convolutional Neural Network (CNN) technique to identify the human and nonhuman image, and the Kalman filter technique to track human movement. For training and testing purposes, there are two kinds of RGB-D datasets used with different points of view and lighting conditions. Both datasets have been selected to remove images which contain a lot of noises and occlusions so that during the training process it will be more directed. Using these integrated techniques, detection and tracking accuracy reach 77.7%. The impact of using Kalman filter increases computation efficiency by 41%.

Full Text Cite


Related Journals

USAbility Testing Of Virtual Keyboard In Touch Screen With Questionnaire Method

The Effect of Lecturer's Professionalism and Student's Study Interest and Motivation on Academic Achievement of Bachelor Degree Student in Online Learning in Binus University

Feature Image Watermarking Based on Bicubic Interpolation of Wavelet Coefficients Using CRT


More

Search Research and Publications

CARI TULISAN is a scientific publication indexing site that helps everyone find research results and relevant data from papers, journals, books, research reports, and so on. Collected from various repositories, it makes scattered scientific research easily searchable.
All articles and content on this site are copyrighted works of the relevant authors that have been published as a result of scientific research. CARI TULISAN never distributes and supports pirated content.