VEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC
Detection and type classification of vehicles are part of Intelligent Transportation System. This paper proposed system for detection and type classification of vehicles based on Oriented FAST and Rotated BRIEF (ORB) method to extract local features in vehicles image, hamming distance for matching l...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/23323 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| id |
id-itb.:23323 |
|---|---|
| spelling |
id-itb.:233232017-09-27T15:37:11ZVEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC RIZQI SHOLAHUDDIN (NIM : 23514074), MUHAMMAD Indonesia Theses INSTITUT TEKNOLOGI BANDUNG https://digilib.itb.ac.id/gdl/view/23323 Detection and type classification of vehicles are part of Intelligent Transportation System. This paper proposed system for detection and type classification of vehicles based on Oriented FAST and Rotated BRIEF (ORB) method to extract local features in vehicles image, hamming distance for matching local features and RANSAC to eliminate matching errors. Evaluation of performance used real vehicle images taken from multiple viewpoints and sizes of google street view. The vehicles are categorized into 5 classes: trucks, buses, sedans, pick-ups, and mpv/suv/van. The results show that the accuracy from 717 test images is 94.28%. From the results, it can be concluded that the use of ORB, hamming distance, and RANSAC algorithm are able to categorize the vehicles more specifically and robustly with changes in viewpoint and size. text |
| institution |
Institut Teknologi Bandung |
| building |
Institut Teknologi Bandung Library |
| continent |
Asia |
| country |
Indonesia Indonesia |
| content_provider |
Institut Teknologi Bandung |
| collection |
Digital ITB |
| language |
Indonesia |
| description |
Detection and type classification of vehicles are part of Intelligent Transportation System. This paper proposed system for detection and type classification of vehicles based on Oriented FAST and Rotated BRIEF (ORB) method to extract local features in vehicles image, hamming distance for matching local features and RANSAC to eliminate matching errors. Evaluation of performance used real vehicle images taken from multiple viewpoints and sizes of google street view. The vehicles are categorized into 5 classes: trucks, buses, sedans, pick-ups, and mpv/suv/van. The results show that the accuracy from 717 test images is 94.28%. From the results, it can be concluded that the use of ORB, hamming distance, and RANSAC algorithm are able to categorize the vehicles more specifically and robustly with changes in viewpoint and size. |
| format |
Theses |
| author |
RIZQI SHOLAHUDDIN (NIM : 23514074), MUHAMMAD |
| spellingShingle |
RIZQI SHOLAHUDDIN (NIM : 23514074), MUHAMMAD VEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC |
| author_facet |
RIZQI SHOLAHUDDIN (NIM : 23514074), MUHAMMAD |
| author_sort |
RIZQI SHOLAHUDDIN (NIM : 23514074), MUHAMMAD |
| title |
VEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC |
| title_short |
VEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC |
| title_full |
VEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC |
| title_fullStr |
VEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC |
| title_full_unstemmed |
VEHICLE DETECTION AND TYPE CLASSIFICATION USING ORB - RANSAC |
| title_sort |
vehicle detection and type classification using orb - ransac |
| url |
https://digilib.itb.ac.id/gdl/view/23323 |
| _version_ |
1822020063537922048 |