Silhouette Object Recognition Using Edge-Based Method
Image processing is very vast domain, in which object recognition is toughest challenge in computer vision. As object having different key feathers to describe them we have taken silhouette image of object from recognition proceeding. In this paper edge d
- PDF / 222,125 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 48 Downloads / 234 Views
Abstract Image processing is very vast domain, in which object recognition is toughest challenge in computer vision. As object having different key feathers to describe them we have taken silhouette image of object from recognition proceeding. In this paper edge detection method is used to detect object in frame and then compare their angle with the database angle value is done. To reduce the size of hardware and speed-up the performance we use Raspberry-Pi 2 model. Keywords Object recognition
Canny Silhouette object Raspberry Pi
1 Introduction Image recognition is an important function in the field of computer vision. Processing on image or video to achieve useful result is very challenging task. Object detection means detecting objects of a certain class (such as books, faces, or cars) in image. Object recognition is a procedure for pointing out an object in an image and tells user about its detail. Object recognition algorithm depends on matching, learning, or pattern recognition algorithms using shape-based or color-based techniques [2, 3]. Recognition algorithm should be accurate and fast enough for real-time applications. Feature selection is one of the most important steps for object recognition. For high accuracy and real-time object recognition, features should be discriminative, robust, and easy to compute. Edge-based algorithm had been widely used features for object detection and object recognition, due to their robustness, simplicity, and speed. Edges are stable object features in the presence of different light intensity and not affected by change in scales [1]. Edge-based algorithm was also used for shape recognition and matching [3, 4]. P. Yadav M.S. Nagmode (&) Department of Electronics & Communication, MITCOE, Pune, India e-mail: [email protected] P. Yadav e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2018 H.S. Saini et al. (eds.), Innovations in Electronics and Communication Engineering, Lecture Notes in Networks and Systems 7, https://doi.org/10.1007/978-981-10-3812-9_26
251
252
P. Yadav and M.S. Nagmode
This paper presents a recognition system by using image processing algorithm as the core element and the cost can be reduced by using a simple webcam. The image processing will be loaded on a board named Raspberry-Pi 2. It is a single-board credit-card-sized system prepared in the United Kingdom (UK) by the Raspberry Pi Foundation [5]. There are two models and both models are similar except for model B+ have the Ethernet, 512 MB SDRAM and 2 USB ports. Both models run on Linux operating system.
2 Background All of us have played the jigsaw puzzle. We have a lot of small pieces of images, where we need to assemble them in correct order to form a big image. How it works? This same theory is projecting to a computer program. So answer is, we are looking for specific unique features, which can be easily tracked and compared. But if someone asks us to point out one good feature which can be compared across several images, we can point out one very easily. We
Data Loading...
