LAND USE CLASSIFICATION USING DEEP MULTITASK NETWORKS
نویسندگان
چکیده
منابع مشابه
Quantitative Toxicity Prediction Using Topology Based Multitask Deep Neural Networks.
The understanding of toxicity is of paramount importance to human health and environmental protection. Quantitative toxicity analysis has become a new standard in the field. This work introduces element specific persistent homology (ESPH), an algebraic topology approach, for quantitative toxicity prediction. ESPH retains crucial chemical information during the topological abstraction of geometr...
متن کاملEvolutionary Architecture Search For Deep Multitask Networks
Multitask learning, i.e. learning several tasks at once with the same neural network, can improve performance in each of the tasks. Designing deep neural network architectures for multitask learning is a challenge: There are many ways to tie the tasks together, and the design choices matter. The size and complexity of this problem exceeds human design ability, making it a compelling domain for ...
متن کاملLand-use Classification Using Sar-images
This paper studies the applicability of ERS-1 and JERS SAR-images to land-use classification. Different factors affecting classification accuracy like environmental conditions of study area, speckle filtering, spatial averaging, texture measures and different ways to use temporal images are discussed. Classification experiment employing three principal component images computed from median filt...
متن کاملLand Use Classification Using Optimum-Path Forest
It was introduced in this paper the Optimum-Path Forest for land use classification aiming a better environmental management, using images obtained from CBERS 2B CCD satellite covering the area of the Rio das Pedras watershed, Itatinga City, São Paulo State, Brazil. We also compared the Optimum-Path Forest algorithm with the well known supervised classifiers: Artificial Neural Networks using Mu...
متن کاملLand-use Classification Using Temporal Sar-images
14 ERS-1 SAR images were classified using three different classification algorithms. Median filtering and principal component analysis were used in speckle reduction and feature extraction. Feature selection was performed using branch-and-bound algorithm. The best classification accuracy (55-60%) was achieved when principal component images computed from median filtered SAR images were used. In...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
سال: 2020
ISSN: 2194-9034
DOI: 10.5194/isprs-archives-xliii-b3-2020-17-2020