A People-Counting System Using a Hybrid RBF Neural Network

ArticleinNeural Processing Letters 18(2):97-113 · October 2003with17 Reads
DOI: 10.1023/A:1026226617974 · Source: DBLP
A people-counting system using hybrid RBF neural network is described. The proposed system is effective and flexible for the purpose of performing on-line people counting. Compared with other conventional approach, this system introduces a novel method for feature extraction. In this Letter, a new type of hybrid RBF network is developed to enhance the classification performance. The hybrid RBF based people-counting system is thoroughly compared with other approaches. Extensive and promising results were obtained and the analysis indicates that the proposed hybrid RBF based system provides excellent people-counting results in an open passage. A supervised clustering method is proposed for initialising the hybrid RBF network. In order to substantiate the introduction of the hybrid RBF and the proposed supervised clustering algorithm, test results on a vowel recognition benchmark dataset are also included in the Letter.
    • "This method can eliminate the deficiency in the mode method. The computing formula of the new method is the following: (1) Here constant Alpha α can be identify by using binaryzation of an image (BW n (x, y)). To get binary image, First calculate distance between current frame and previous frame that can compare with threshold value. "
    [Show abstract] [Hide abstract] ABSTRACT: The paper presents an efficient and reliable approach to automatic people segmentation, tracking and counting, used in surveillance systems. The initial novel background extraction algorithm uses an improved mode algorithm to obtain the static and novel fuzzy background subtraction approach regions for background subtraction. We also develop a complete framework to evaluate a new edge flow vector based segmentation processes. Tracking segmented people is a dynamic cluster assignment problem between two consecutive frames, and it is solved by a fuzzy-based rule system for tracking and people counting and is applied to people surveillance. Experimental results suggest that the proposed method can achieve good results in both counting accuracy and execution speed.
    Full-text · Article · Dec 2015
    • "Like in [1], a vision based technique is reported for detection of presence of object. A hybrid RBF network based online peoplecounting system is discussed in [2]. A traffic management system with audio and video data at intersections is reported in [3] [4] for removal of stromal cells from breast histopathology images using segmentation and classification techniques. "
    [Show abstract] [Hide abstract] ABSTRACT: Modernization in civilization is making way for demand of school gears day by day. To balance consumption it is very essential to increase production. Its often found that incorporation of automotive technologies in plants would enhance production. In view of these prevailing conditions, proposed work presents an automated technique for (i) detection of pencil color and further classify the pencil based on color, (ii) perform quantitative analysis on pencil packet, to check whether desired number of pencils are present. Whole technique is designed using non contact technique for quicker and undisturbed flow of products on conveyor. A high resolution camera is placed along to conveyor carrying pencils post production. Acquired image is processed using image processing and neural network algorithms for identification, classification of pencils and quantitative analysis of pencils is performed on LabVIEW platform. To demonstrate working of proposed technique a conveyor mechanism is designed. Proposed technique is tested on real life setup, results produced shows successful implementation of set objectives.
    Full-text · Article · Jan 2015 · Discrete Dynamics in Nature and Society
    • "Most of the previous works can be classified into two classes by camera view. One captures the video clips by longshot and nondirect downward view camera [1] [2] [3] [4] [5] and the other by short-shot and direct downward view camera [6] [7] [8] [9] [10] [11] [12] [13] (see Figure 1). The image-processing method of the first class is the first to detect and track the foreground group by body shape feature and then segment the group into individuals for counting. "
    [Show abstract] [Hide abstract] ABSTRACT: A pedestrian counting method based on Haar-like detection and template-matching algorithm is presented. The aim of the method is to count pedestrians that are in a metro station automatically using video surveillance camera. The most challenging problem is to count pedestrians accurately in the case of not changing the position of the surveillance camera, because the view that surveillance camera uses in a metro station is always short-shot and nondirect downward view. In this view, traditional methods find it difficult to count pedestrians accurately. Hence, we propose this novel method. In addition, in order to improve counting accuracy more, we present a method to set the parameter value with a threshold-curve instead of a fixed threshold. The results of experiments show the high accuracy of our method.
    Full-text · Article · Feb 2014
Show more