Abstract—The security of restricted areas such as borders or buffer zones is of utmost importance; in particular with the worldwide increase of military conflicts, illegal immigrants, and terrorism over the past decade. Monitoring such areas rely currently on technology and man power, however automatic monitoring has been advancing in order to avoid potential human errors that can be caused by different reasons. This paper introduces an automatic moving object detection, extraction and recognition system (aMODERs), which uses image processing to detect and extract moving objects within a restricted area, and a neural network to recognize the extracted object. The proposed system monitors movement by humans, animals or vehicles across a secured zone. Experimental results indicate that (aMODERs) provides a simple, efficient and fast solution to the problem of detecting, extracting and recognizing moving objects within one system. Keywords— Moving object detection, extraction and recognition, neural network classification, security systems. I. INTRODUCTION OVING object detection is a basic and important problem in video analysis and vision applications. Automatic recognition systems for still and moving objects can be invalid in security applications, such as monitoring border areas, buffer zones and restricted areas. A simple recognition system would comprise a camera fixed high above the monitored zone, where images of the zone are captured and consequently processed. Processing the captured images can be in three phases, namely, detection of a moving object, extraction of the object and finally recognition of the object. Optical flow and background subtraction have been used for detecting moving objects in image sequences [1]-[4]. For example, adaptive optical flow for person tracking [2] is dependent on being able to locate a person accurately across a series of frames. Optical flow can be used to segment a moving object from a scene, provided the expected velocity of the moving object is known; but successful detection also relies on being able to segment the background. Other works such as moving objects segmentation using optical flow Manuscript received Febr. 8, 2008; Revised version received April 4, 2008. A. Khashman is with the Intelligent Systems Research Group (ISRG), Department of Electrical & Electronic Engineering, Near East University, Nicosia, Cyprus (corresponding author, phone: +90-392-223-6464; fax: +90- 392-223-6622; e-mail: amk@neu.edu.tr). estimation [3] presented a method for the segmentation of moving objects, where a powerful variation method using active contours for computing the optical flow is used. However, the high computational time to extract the optical flow and the lack of discrimination of the foreground from the background, make this method unsuitable for real time processing. On the other hand, background subtraction detects moving objects by subtracting estimated background models from images. This method is sensitive to illumination changes and small movement in the background, e.g. leaves of trees [4]. Another problem of background subtraction is that it requires a long time for estimating the background models. It usually takes several seconds for background model estimation because the speed of illumination changes and the small movement in the background are very slow. Much research work into moving object recognition has been previously presented. In [5], a semantic video object generation and temporal tracking technique for providing content-based video representation and indexing were proposed. In [6], a Gramian determinant-based method was proposed to detect moving objects between multiple images and to detect changes between color images or any type of multi-spectral images. In [7], a moving object detection method was proposed using global motion estimation and edge information, where the final objects are extracted by combining the contours and moving regions from motion detection. In [8], a method based on background subtraction was proposed to recognize abnormal human behavior in public areas using segmentation of moving objects in real- time from images acquired by a fixed color video camera. In [9], a system for moving object detection and shadow extermination by building an adaptive background model was described. In [10], a method for detecting poorly visible moving objects in bad weather was proposed, and was based on measuring a variation on a cross correlation between a short accumulated histogram and along accumulated histogram of detected objects. In [11], a method was proposed to extract contours of moving objects mainly by combining gradient information extracting with three-frame differencing and connectivity-testing-based noise reduction. In [12], a binary image filtering method based on discrete information entropy was proposed for moving object detection in image sequences. In [13], a method for detecting and extracting moving objects from a moving stereo camera was described, where camera motions were calculated from three- dimensional optical flow, and the moving object was detected Automatic Detection, Extraction and Recognition of Moving Objects Adnan Khashman M INTERNATIONAL JOURNAL OF SYSTEMS APPLICATIONS, ENGINEERING & DEVELOPMENT Issue 1, Volume 2, 2008 43