System Architecture for Real-Time Face Detection on Analog Video Camera

Abstract

This paper proposes a novel hardware architecture for real-time face detection, which is efficient and suitable for embedded systems. The proposed architecture is based on AdaBoost learning algorithm with Haar-like features and it aims to apply face detection to a low-cost FPGA that can be applied to a legacy analog video camera as a target platform. We propose an efficient method to calculate the integral image using the cumulative line sum. We also suggest an alternative method to avoid division, which requires many operations to calculate the standard deviation. A detailed structure of system elements for image scale, integral image generator, and pipelined classifier that purposed to optimize the efficiency between the processing speed and the hardware resources is presented. The performance of the proposed architecture is described in comparison with the detection results of OpenCV using the same input images. For verification of the actual face detection on analog cameras, we designed an emulation platform using a low-cost Spartan-3 FPGA and then experimented the proposed architecture. The experimental results show that the processing time for face detection on analog video camera is 42 frames per second, which is about 3 times faster than previous works for low-cost face detection.

Mask Detection Analysis Using Haar Cascade and Naïve Bayes

“…Then, Viola and Jones did a development that led to the creation of the Haar-Like feature. The Haar-Like feature processes images in the form of boxes, which in a box consists of a number of pixels, each square which is then processed and looks for differentiating values that indicate dark areas and bright areas [21]. These values will then be the basis for image processing [22].…”

A Machine Learning-Based Approach for Counting Blister Cards Within Drug Packages

“…There are a lot of applications for object detection algorithms. In [28], a novel method which is able to detect the face in analog cameras was proposed [28]. First, an efficient method for calculation of integral image via using of cumulative line sum was implemented then, in order to avoid division, an alternative approach found on compact integral image generator was suggested.…”

“…The normalization factor was designed to show the average intensity in a feature region. A new model for object detection was evaluated in [28]. In that method, several innovative ideas for Adaboost learning was presented.…”

“…In order to determine if the subwindow contains an object, the candidate should pass all the stages in the cascade. To implement Haar cascade classifier, we used OpenCV libraries [28].…”

An image processing-based system proposal for real-time detection of drowsiness from a vehicle driver's eye movements

“…On the other hand, most commercially available contactless driver drowsiness detection systems and related products use visible light to achieve face detection [4]. During the day, they do not affect drivers; nevertheless, at night, since these systems need to gather the skin-color region, they are likely to increase the burden on driver's eyes since extra light was projected on the driver's face to properly obtain color images [5].…”