An Object Recognition and Identification System Using the Hough Transform Method

Abstract

Abstract-This paper presents an object recognition and identification system using the Hough Transform method. The process starts from imported images into the system by webcam, detected image edge by fuzzy, recognized the object by Hough Transform, and separated the objects by the robot arm. Three objects type; triangle, rectangular and, rigid circle are used. The results showed that the objects can be isolated 96%, 96%, and 98% correct for triangular, rectangular, and rigid circle respectively. Index Terms-Hough transform, modifier fuzzy sobel, object recognition I. INTRODUCTIONNowadays, the robots are used widely in many applications such as automotive industry, rescue and security areas, medical exploration robotic, etc. Manipulator arms are employed in the industrial as the human arm for welding, lifting, separating, etc. Sometimes referred to as robot arm also means that the robot in industry. The robots will be played greater roles in the industry as much as the robot control algorithms are developed. They will work in various hazardous jobs such as lifting steel into furnace, job-related chemicals, repetitive monotonous work such as lifting or packing the objects in production line, desired quality such as welding and cutting, or the jobs that required highly skills such as welding line, welding laser, etc.In this research, we are presented an object recognition and identification system using the Hough Transform method. There are many researches that used the Hough Transform method. Basak and Pal generalize the classical Hough Transform in fuzzy set theoretic framework in order to handle the imprecise in shape description called fuzzy Hough transform [1]. Maji and Malik present a discriminative Hough transform based object detector where each local part casts a weighted vote for the possible locations of the object center. The weights can be learned in a max-margin framework which directly optimizes the classification performance and improved the Hough detector [2]. Rizon and others use the circular Hough transform to detect the presence of circular shape [3]. Smereka and Duleba used The Hough Transform to improve the detection of low-contrast circular objects [4]. The Hough Transform can be implemented in C as discussed by Lee [5]. It also was modify to reduce memory area, computational time, and detect noise which was present by Sirisantisamrid and et al [6]. Khoshelham used laser range data and modified the generalized Hough Transform to detect 3D objects [7]. Alshennawy and Aly introduce the fuzzy logic reasoning to detect the digital images edge [8]. The linear Sobel operator was used to give a permanent effect in the lines smoothness and straightness for the straight lines and good roundness for the curved lines.In our work, a robot arm was designed to use a fuzzy logic to detect the image edge [8] and to use Hough Transform method to recognize, identify, and separate the objects. The triangle, rectangular, and rigid circle objects are used to verify the system. II. IMAGE PREPROCESSING AND ENHANCE...