Loading...

Vision-based Vehicle Detection in Intercity Roads for Intelligent Transportation Systems Applications

Rostami, Peyman | 2018

497 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 51742 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Marvasti, Farokh
  7. Abstract:
  8. This project aims to highlight vision related tasks centered around "car". First, we gathered a dataset of 4343 front view car images, captured from the streets of Iran and Syria during daylight, the images of which are all manually cropped around their corresponding accurately chosen bounding boxes. we also extracted seven parts (i.e. left and right front lights, left and right mirrors, bumper, plate, and air intake) from each car image in the dataset. Our dataset is suitable for developing and testing bounding box extraction algorithms, holistic and part based analyses, occlusion handling algorithms, etc. next, we utilized Viola-Jones Detector to develop a system for car detection, in which we proposed a method for extracting a quasi-confidence measure from the attentional cascade of independently trained Adaboost classifiers. Later on, the quasi-confidence measure was used not only for bounding box extraction in a set of overlapping multiple detections, using none maximum suppression, which resulted in an alignment accuracy of 87% but also for increasing our systems detection accuracy in terms of false positive and false negative rates, the promising results of which were corroborated by improvements in our system's ROC curve
  9. Keywords:
  10. Machine Vision ; Machine Learning ; Database ; Vehicle Detection ; Intelligent Transportation System (ITS) ; Viola-Jones Object Detection

 Digital Object List

 Bookmark

No TOC