14802

13322

كارشناسي ارشد

الكترونيك

اصفهان : دانشگاه صنعتي اصفهان

1398

سيزده، 103ص.: مصور، جدول، نمودار

1398/05/02

كتابنامه

برق و كامپيوتر

مهندسي برق و كامپيوتر

1398/05/02

2549772

Hardware Implementation of Object Detection in Image Using Histogram of Oriented Gradients and Support Vector Machine Parisa Amjadian Amjadianp@gmail com Date of Submission 2019 5 20 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language FarsiSupervisor Dr Sayed Masoud Sayedi m sayedi@cc iut ac irAbstractObject recognition in the image due to its many applications like in human recognition detection of cancercells and fingerprint recognition is an important task in image processing Various methods have beenproposed for object detection among which Histogram of Oriented Gradients HOG as a special tool forfeature extraction has a lot of importance This tool has better performance compared with other tools interms of response to illumination object rotation and shape variations Also Support Vector machine SVM is a powerful classification tool for large dimensional input data in various image processingapplications In this method for the classification of the data the best hyperplane with the largest margin andleast classification error is selected This method can be used for a variety of applications because of theexistence of many different kernel functions Real time detection of an object plays an important role inmany applications such as military medical intelligent transportation monitoring security entertainmentand robotics applications So designing a system that can handle the required processing speeds is veryimportant and the higher speeds of hardware systems compared to the software ones has dramaticallyincreased the tendency to use hardware based detection algorithms In present work first a survey on theHOG and SVM methods is provided and related implementations are studied Then a hardwareimplementation of object detection system based on these methods is presented Keywords Hardware implementation object detection image processing histogram of oriented gradients supportvector machine

مسعود سيدي

نادر كريمي، مازيار پالهنگ