Int. J. Advanced Networking and Applications 956 Volume: 02, Issue: 06, Pages: 956-962 (2011) Automatic Change Detection Method for an Indoor Environment D. Beulah David Department of Computer Science, Jeppiaar Engineering College, Chennai-119 Email: beulahchrist@gmail.com -------------------------------------------------------------------ABSTRACT-------------------------------------------------------------- Detecting changed regions in multiple images of the same scene taken at different times is of great interest due to a large number of applications in diverse disciplines, including remote sensing, surveillance, medical diagnosis and treatment, civil infrastructure, and underwater sensing. The goal is to identify the set of pixels that are significantly different between the last image of the sequence and the previous images. This paper presents a study of an automatic change detection method for multiple images of the same scene acquired by a mobile-camera from different positions with no illumination changes. The proposed method used for this study consists of three steps: (1) automatic image registration, (2) temporal differencing and (3) unimportant changes removal. The results in this paper show that the presented method successfully detects new objects in the scene from the multiple images. Keywords  Automatic Image Registration, Temporal differencing , Unimportant Changes Removal. ------------------------------------------------------------------------------------------------------------------------------------------------------ Date of Submission: 11 November 2010 Date of Acceptance: 26 April 2011 ------------------------------------------------------------------------------------------------------------------------------------------------------ 1. INTRODUCTION A major portion of the research efforts of the computer vision community has been directed towards the study of automatic image change detection methods [1] due to a large number of applications in diverse disciplines. Important applications of change detection include video surveillance, analysis of multitemporal remote sensing images, underwater sensing [2], tracking systems of moving objects, medical diagnosis, or driver assistance systems [3]. Given a set of images of the same scene taken at several different times, the goal is to identify the set of pixels that are significantly different between the last image and a previous reference image. Changed pixels may result from a combination of underlying factors, including appearance or disappearance of objects, motion of objects relative to the background, shape change of objects or environment modifications (buildings, fires), etc., [1]. Change detection methods applied to images of the same scene captured by stationary cameras at different times are widespread, both in the literature [8] and in everyday use, but only few publications refer to image analysis from non-stationary camera positions. Oberti et al. [6] presented a video surveillance system based on a pan/tilt mobile-camera. Their method began by creating a panoramic multi-layer background image. The background image and the camera position when capturing an input image were used to perform change detection. While the system could detect a new object in a video sequence from a non-static camera, the system depended heavily on the current position of the camera when capturing an input image. When this position of the camera was unknown, the system failed to detect changes in the input image. Primdahl et al. [7] described an approach that analyzes videos acquired from a moving vehicle making repeated passes through a specific, well defined corridor. The objective of their approach was to detect stationary objects that appear in the scene along the established route. Their approach began with pairing images of different videos. For each frame pair, they introduced regions of interest. Although the new objects were successfully identified out of the videos captured from different camera locations, this only occurred because the authors manually selected four points on the regions of interest that contained the new objects. The authors explicitly identified the locations of the new objects. 2. TECHNIQUES IN IMAGE CHANGE DETECTION 2.1. Image Change Detection: Types 1) Temporal Difference Models 2) Vector and Shading Models 3) Clustering Models 4) Significance and Hypothesis Tests Models 2.2. Contribution of This Research In this paper, an automatic change detection method is presented for multiple indoor images of the same scene captured by a mobile-camera from different positions. The goal of the proposed system is to detect new objects present in the images. The system does not require information of mobile-camera positions and regions of interest as input. In future, the proposed change detection method could be applied to patrol large areas such as perimeter