Despite all the significant advances in human detection in various environmental conditions, it is still a challenging task. Most of the human detection algorithms mainly use color information, which is not robust to lighting changes and varying colors under which such a detector should operate namely day and nighttime. This problem is further amplified with infrared (IR) imagery, which only contains grayscale information. The proposed algorithm for human detection uses intensity distribution, gradient, and texture features for effective detection of humans in IR imagery. For the detection of intensity, histogram information is obtained in the grayscale channel. For extracting gradients, we utilize the histogram of oriented gradients for better information in the various lighting scenarios. For extracting texture information, center-symmetric local binary pattern gives rotational invariance as well as lighting invariance for robust features under these conditions. Various binning strategies help keep the inherent structure embedded in the features, which provide enough information for robust detection of the humans in the scene. The features are then classified using an AdaBoost classifier to provide a tree-like structure for detection in multiple scales. The algorithm has been trained and tested on IR imagery and has been found to be fairly robust to viewpoint changes and lighting changes in dynamic backgrounds and visual scenes.
Many human detection algorithms are able to detect humans in various environmental conditions with high accuracy, but they strongly use color information for detection, which is not robust to lighting changes and varying colors. This problem is further amplified with infrared imagery, which only contains gray scale information. The proposed algorithm for human detection uses intensity distribution, gradient and texture features for effective detection of humans in infrared imagery. For the detection of intensity, histogram information is obtained in the grayscale channel. For extracting gradients, we utilize Histogram of Oriented Gradients for better information in the various lighting scenarios. For extraction texture information, center-symmetric local binary pattern gives rotational-invariance as well as lighting-invariance for robust features under these conditions. Various binning strategies help keep the inherent structure embedded in the features, which provide enough information for robust detection of the humans in the scene. The features are then classified using an adaboost classifier to provide a tree like structure for detection in multiple scales. The algorithm has been trained and tested on IR imagery and has been found to be fairly robust to viewpoint changes and lighting changes in dynamic backgrounds and visual scenes.
In this paper we present a low level image descriptor called Histogram of Oriented Phase based on phase congruency concept and the Principal Component Analysis (PCA). Since the phase of the signal conveys more information regarding signal structure than the magnitude, the proposed descriptor can precisely identify and localize image features over the gradient based techniques, especially in the regions affected by illumination changes. The proposed features can be formed by extracting the phase congruency information for each pixel in the image with respect to its neighborhood. Histograms of the phase congruency values of the local regions in the image are computed with respect to its orientation. These histograms are concatenated to construct the Histogram of Oriented Phase (HOP) features. The dimensionality of HOP features is reduced using PCA algorithm to form HOP-PCA descriptor. The dimensionless quantity of the phase congruency leads the HOP-PCA descriptor to be more robust to the image scale variations as well as contrast and illumination changes. Several experiments were performed using INRIA and DaimlerChrysler datasets to evaluate the performance of the HOP-PCA descriptor. The experimental results show that the proposed descriptor has better detection performance and less error rates than a set of the state of the art feature extraction methodologies.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.