A Novel Siamese-Based Approach for Scene Change Detection With Applications to Obstructed Routes in Hazardous Environments
| Autor: | Thierry Pinheiro Moreira, Leandro Aparecido Passos Junior, João Paulo Papa, Victor Hugo C. de Albuquerque, Danilo Colombo, Marcos C. S. Santana |
|---|---|
| Přispěvatelé: | Universidade Estadual Paulista (Unesp), Petr Brasileiro SA Petrobras, Univ Fortaleza |
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
Computer Networks and Communications
Computer science Siamese Convolutional Neural Networks Decoding Scene Change Detection 02 engineering and technology Route Obstruction Detection Field (computer science) Artificial Intelligence Robustness (computer science) U-Nets 0202 electrical engineering electronic engineering information engineering Intelligent systems Computer vision Image segmentation Training data business.industry Deep learning Intelligent decision support system Object (computer science) Semantics Task analysis 020201 artificial intelligence & image processing Artificial intelligence Noise (video) Human computer interaction business Change detection Neural networks |
| Zdroj: | Web of Science Repositório Institucional da UNESP Universidade Estadual Paulista (UNESP) instacron:UNESP |
| Popis: | Made available in DSpace on 2020-12-11T14:35:48Z (GMT). No. of bitstreams: 0 Previous issue date: 2020-01-01 Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) Petrobras The demand for automatic scene change detection has massively increased in the last decades due to its importance regarding safety and security issues. Although deep learning techniques have provided significant enhancements in the field, such methods must learn which object belongs to the foreground or background beforehand. In this article, we propose an approach that employs siamese U-Nets to address the task of change detection, such that the model learns to perform semantic segmentation using background reference frames only. Therefore, any object that comes up into the scene defines a change. The experimental results show the robustness of the proposed model over the well-known public dataset CDNet2014. Additionally, we also consider a private dataset called PetrobrasROUTES, which comprises obstruction or abandoned objects in escape routes in hazardous environments. Moreover, the experiments show that the proposed approach is more robust to noise and illumination changes. Sao Paulo State Univ, Sao Paulo, Brazil Petr Brasileiro SA Petrobras, Rio De Janeiro, Brazil Univ Fortaleza, UNIFOR, Fortaleza, Ceara, Brazil Sao Paulo State Univ, Sao Paulo, Brazil FAPESP: 2013/07375-0 FAPESP: 2014/12236-1 FAPESP: 2016/19403-6 FAPESP: 2017/25908-6 CNPq: 307066/2017-7 CNPq: 427968/2018-6 CNPq: 304315/2017-6 CNPq: 430274/2018-1 Petrobras: 2017/00285-6 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|