A new decision fusion technique for image classification

2009 ◽  
Author(s):  
Mete Ozay ◽  
Fatos Tunay ◽  
Yarman Vural
Keyword(s):  
Image Classification ◽  
Decision Fusion ◽  
Fusion Technique

scholarly journals Dermoscopy Image Classification Based on StyleGANs and Decision Fusion

IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 70640-70650 ◽  
Author(s):  
An Gong ◽  
Xinjie Yao ◽  
Wei Lin
Keyword(s):  
Image Classification ◽  
Decision Fusion

scholarly journals Hyperspectral Imagery Classification Based on Multiscale Superpixel-Level Constraint Representation

2020 ◽  
Vol 12 (20) ◽  
pp. 3342
Author(s):  
Haoyang Yu ◽  
Xiao Zhang ◽  
Meiping Song ◽  
Jiaochan Hu ◽  
Qiandong Guo ◽  
...  
Keyword(s):  
Image Classification ◽  
Satellite Data ◽  
Hyperspectral Image ◽  
Training Sample ◽  
Decision Fusion ◽  
Spatial Constraint ◽  
Hyperspectral Image Classification ◽  
Object Based ◽  
Sparse Coefficient ◽  
The Individual

Sparse representation (SR)-based models have been widely applied for hyperspectral image classification. In our previously established constraint representation (CR) model, we exploited the underlying significance of the sparse coefficient and proposed the participation degree (PD) to represent the contribution of the training sample in representing the testing pixel. However, the spatial variants of the original residual error-driven frameworks often suffer the obstacles to optimization due to the strong constraints. In this paper, based on the object-based image classification (OBIC) framework, we firstly propose a spectral–spatial classification method, called superpixel-level constraint representation (SPCR). Firstly, it uses the PD in respect to the sparse coefficient from CR model. Then, transforming the individual PD to a united activity degree (UAD)-driven mechanism via a spatial constraint generated by the superpixel segmentation algorithm. The final classification is determined based on the UAD-driven mechanism. Considering that the SPCR is susceptible to the segmentation scale, an improved multiscale superpixel-level constraint representation (MSPCR) is further proposed through the decision fusion process of SPCR at different scales. The SPCR method is firstly performed at each scale, and the final category of the testing pixel is determined by the maximum number of the predicated labels among the classification results at each scale. Experimental results on four real hyperspectral datasets including a GF-5 satellite data verified the efficiency and practicability of the two proposed methods.

A comprehensive Overview of Decision Fusion Technique in Healthcare: A Systematic Scoping Review

2020 ◽  
Keyword(s):  
Health Care ◽  
Health Informatics ◽  
Holistic Approach ◽  
Decision Fusion ◽  
Cost Of Care ◽  
Management Technique ◽  
Care Providers ◽  
Comprehensive Overview ◽  
Fusion Technique

Introduction: Decision fusion has emerged as a data management technique due to the diversity and scalability of data in health care. This first-scope review aimed to investigate the use of this technique in health care. Materials and Methods: A query was carried out on PubMed, Science Direct, and EMBASE within 1960-2017 using such keywords as decision fusion, information fusion, symbolic fusion, distributed decisions, expert fusion, and sensor fusion, in conjunction with med-* and health-care. The articles were analyzed in terms of methodology and results. Results: The literature search yielded 106 articles. Based on the results, in the field of health care, the articles were related to image processing (29%), sensors (22%), diagnosis area(10%), biology (6%), health informatics (8%), and signal process (15%). The majority of articles were published in 2011, 2012, and 2015, and the USA had the largest number of articles. Most of the articles were about engineering and basic sciences. Regarding healthcare, the majority of studies were conducted on the diagnosis of diseases (80%), while 9% and 11% of articles were about prevention and treatment, respectively. These studies applied the following methods: intelligent methods (44%), new methods (36%), probabilistic (13%), and evidential methods (7%). The dataset was as follows: research project data (49%), online dataset (42%), and simulation (9%). Furthermore, 49% of articles mentioned the applied software, among which classification and interpretation were reportedly the most and the least used methods. Discussion and Conclusion: Decision fusion is a holistic approach to evaluate all areas of health care and elucidate diverse techniques that can lead to improved quality of care. Innovation: This article is the first scope review article about the application of the decision fusion technique in the field of health care, building on an established protocol. Decision fusion can reduce the cost of care and improve the quality of health care provision. Therefore, this article can help care providers understand the benefits of this technique and overcome challenges in adopting decision fusion technology.

Sign in / Sign up

Export Citation Format

Share Document