International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-9 Issue-2, December, 2019 2538 Retrieval Number: B3830129219/2019©BEIESP DOI: 10.35940/ijeat.B3830.129219 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Scene Illustration of Terrestrial Animals with Its Monitoring, Tracking and Recognizing Through Deep Learning in Relation with Granular Computing Neelam Rawat, J.S. Sodhi, Rajesh K. Tyagi Abstract: Combining Deep Learning Technique with Granular Computing employs an inductive paradigm for the terrestrial animal’s elucidation. The proposed method frames the object (terrestrial animal) in arbitrary-shaped and sized granules rather than fixed and rectangular shaped, so that object can effectively mine and recognized. The goal is to present a formal model which automatically focus only on representative pixel of each granule rather than converting pixels from entire image through scanning. Thus, this work entails the process of recognizing not only the static animal in the background, but also depicts moving animal in foreground separately. Keywords: Granular Computing (GrC), Deep Learning, Object Recognition, Object Tracking, CNN. GPRS I. INTRODUCTION From last decades, camera-trap images are the most powerful tool to evaluate and automatically extract the activities (e.g., eating, sleeping, running, drinking) which provides a valuable information to ecological studies [4]. In this paper, we extract to assess a specific hypothesis data that enable future ecological studies to have detailed and large-scale knowledge about the terrestrial animal movements in natural ecosystems that is even challenging for humans. In the monitoring and tracking process, frame rate goes high when detection process for target animal is with consecutive video frames. Even though complexity increases with addition of time and object orientation in tracking process [5]. The method of using both deep learning and granular computing can automate the process of monitoring and tracking of animals that dramatically improves the detection from those frames (granules). A granule represents the subset of Universe of Discourse (UoD) which is composed of finer granules plotted by the functionality, similarity and distinguishability. Closeness and cohesion of any subset of objects can depict it as a granule. The concept of Granular Computing is to deal with uncertainty, imprecision, partial truth [7]. Revised Manuscript Received on December 15, 2019. Neelam Rawat, Research Scholar, Amity University, Noida, India. Email: neemarawat11@gmail.com J. S. Sodhi, Amity University. Noida, India Email: jssodhi@akcgroup.com Rajesh K. Tyagi, Computer Science, Amity University, Gurugram, India. Email: profrajeshkumartyagi@gmail.com The process of representation and formation of granules called as Granulation (an operation performed on granules). Based on the application, granules can be fuzzy, rough fuzzy or crispy. [17]Process of decomposition makes larger granules into smaller or lower-level granules. Construction in granulation is bottom-up entrance or interrelationship whereas decomposition in granulation is top-down entrance or intra-relationship [3]. Granular Computing (GrC) models Uncertainty which may further categorize as Objective Uncertainty and Subjective Uncertainty (Figure 1.) Figure 1: Dimensions of Uncertainty  Models for subjective uncertainty  Fuzziness – fuzzy sets  Models for objective uncertainty  Randomness – probability theory  Roughness – rough sets  Greyness – grey systems Here, we have Objective Uncertainty, so we use rough set Granular Computing with Deep Learning Technique for recognition of animals and its motion detection [10]. The rest of the paper uphold with several sections as section 2 defines the fundamental understanding of utilized terms and techniques in this work. Section 3 extant the concept of deep learning concepts and the fusion of granular computing with deep learning as granulated deep learning exemplify in section 4. Proposed methodology and its result mentioned in section 5 and section 6 respectively followed be conclusions and references. II. GRANULATION TECHNIQUE In the proposed formal model, granular knowledge appears to focus on the representation of activities performed by animals with various activities likewise sleeping, eating, grazing, standing, waking and so on. Such granular frames, ascribes to be as granular models that quantify the source diversity of granular knowledge and reflect it as a higher order models.