Vehicle detection using morphological shared-weight neural network in the multiple instance learning framework
Abstract
In this thesis, we design and implement an algorithm for object detection in aerial images based on the morphological shared-weight neural network (MSNN). The multiple instance learning (MIL) framework is used to avoid the labeling problem required in a supervised learning framework. Using the MIL, each image was given a single label. We rely on the MSNN's ability to detect objects, and on the methodology used to generate bags to find our target. Two multiple-instance MSNN structures are developed. The performance of this framework is compared with the performance of a convolutional neural network (CNN) in the same condition.
Degree
M.S.
Thesis Department
Rights
OpenAccess.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.