Indonesian Journal of Electrical Engineering and Computer Science Vol. 12, No. 2, November 2018, pp. 783~793 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v12.i2.pp783-793  783 Journal homepage: http://iaescore.com/journals/index.php/ijeecs Object Recognition Inspiring HVS Mohammadesmaeil Akbarpour, Nasser Mehrshad, Seyyed-Mohammad Razavi Electrical Engineering, University of Birjand, Birjand, Iran Article Info ABSTRACT Article history: Received Apr 9, 2018 Revised May 20, 2018 Accepted Jul 11, 2018 Human recognize objects in complex natural images very fast within a fraction of a second. Many computational object recognition models inspired from this powerful ability of human. The Human Visual System (HVS) recognizes object in several processing layers which we know them as hierarchically model. Due to amazing complexity of HVS and the connections in visual pathway, computational modeling of HVS directly from its physiology is not possible. So it considered as a some blocks and each block modeled separately. One models inspiring of HVS is HMAX which its main problem is selecting patches in random way. As HMAX is a hierarchical model, HMAX can enhanced with enhancing each layer separately. In this paper instead of random patch extraction, Desirable Patches for HMAX (DPHMAX) will extracted. HVS for extracting patch first selected patches with more information. For simulating this block patches with more variance will be selected. Then HVS will chose patches with more similarity in a class. For simulating this block one algorithm is used. For evaluating proposed method, Caltech 5 and Caltech101 datasets are used. Results show that the proposed method (DPMAX) provides a significant performance over HMAX and other models with the same framework. Keywords: Object recognition HMAX HVS Hierarchical model Copyright © 2018 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Mohammadesmaeil Akbarpour Electrical Engineering, University of Birjand, Birjand, Iran. Email: esmaeil.akbarpour@birjand.ac.ir 1. INTRODUCTION The Human Visual System (HVS) is able to recognize objects easily in a cluttered scene in less than a second. Recent works inspiring the Human Visual System (HVS) in image processing are such as image enhancement [1], data hiding [2, 3], digital image fusion [4], robust object recognition [5]. HVS processes images easily, while the most powerful computer systems are generally not capable of doing so. Due to the tremendous complexity of HVS and amazing connections in visual pathway, computational modeling of HVS for image processing applications directly from its overall anatomy and physiology is not possible [6]. One of way to overcome the limitation is the input-output modeling of the visual system (i.e. the saliency map) [7, 8]. Another way is modeling of the simple subsystems and their systematically combination based on the HVS structure (i.e. edge and line detection, contour extraction and texture diagnose) [9, 10]. It seems that the manner of the HVS in the object description stage and object recognizing (matching) process is optimized. In the first step of modeling the visual system behavior in object recognition, an appropriate object descriptor should be presented. This descriptor must be independent to scale and rotation [5, 11, 12]. HVS for object description uses processes such as saliency map, edge detection, line detection, contour extraction and texture diagnose. The saliency map [7, 8] is the first topographically arranged map that represents visual saliency of a corresponding visual scene. For edges detection [13, 14] the retina and LGN cells are inspired. They don’t have directional selection because of the circular receptive field.