Bhaskar, Hariyaran
(2020)
Drone Based People Counter And Tracking Using Unique Id Using Opencv Python.
Project Report.
Universiti Sains Malaysia, Pusat Pengajian Kejuruteraan Aeroangkasa.
(Submitted)
Abstract
The purpose of this thesis is to implement an algorithm for drone-based square boundaries using Python and OpenCV. This thesis provided a detailed method in implementing the algorithm for face and person recognition by using haar cascade classifier feature detection and contour approximation. The algorithm is developed so that it is capable to detect various objects by applying a bounding box on the frame image. This project uses Python as its programming language and OpenCV as an open-source library for programming. The image is taken from the DJI Tello drone. The acquired image is then converted to grayscale. Gaussian filter is used for image smoothing and noise removal. The canny edge detector is used for the recognition of an object's edges. Upon implementation, the contours are performed for further analysis and recognition of the person shape. The crosshairs are drawn on the frame for aiming purposes. The testing is done on different images’ characteristics to verify the required features and the problems come out. Also, the experiments are done by using the various value of epsilon to estimate the accuracy of the detection. The angle of projection and the distance between the drone and the object are included in observation. The outcome of this project reflects the object detection technique which will potentially improve the machine vision and will subsequently contribute to the development of the image processing in artificial intelligence.
Actions (login required)
 |
View Item |
