A Comparison of Haze Removal Algorithms and Their Impacts on Classification Accuracy for Landsat Imagery

Abstract

Abstract:The quality of Landsat images in humid areas is considerably degraded by haze in terms of their spectral response pattern, which limits the possibility of their application in using visible and near-infrared bands. A variety of haze removal algorithms have been proposed to correct these unsatisfactory illumination effects caused by the haze contamination. The purpose of this study was to illustrate the difference of two major algorithms (the improved homomorphic filtering (HF) and the virtual cloud point (VCP)) for their effectiveness in solving spatially varying haze contamination, and to evaluate the impacts of haze removal on land cover classification. A case study with exploiting large quantities of Landsat TM images and climates (clear and haze) in the most humid areas in China proved that these haze removal algorithms both perform well in processing Landsat images contaminated by haze. The outcome of the application of VCP appears to be more similar to the reference images compared to HF. Moreover, the Landsat image with VCP haze removal can improve the classification accuracy effectively in comparison to that without haze removal, especially in the cloudy contaminated area. Keywords: haze contamination; HF; VCP; classification accuracy Resumo:A qualidade de imagens Landsat em áreas úmidas é consideravelmente degradada pela neblina em termos do seu padrão de resposta espectral, o que limita a possibilidade da sua aplicação usando as bandas do visível e infravermelho próximo. Uma variedade de algoritmos de remoção de neblina têm sido propostos para corrigir estes efeitos indesejados da iluminação, causados pela contaminação por neblina. O propósito deste estudo foi demonstrar a diferença entre os dois principais algoritmos (a filtragem homomórfica melhorada (HF) e a nuvem virtual de pontos (VCP)), considerando sua efetividade na solução da contaminação da neblina com variação espacial, e para avaliar os impactos da remoção da neblina na classificação da cobertura do solo. Um estudo de caso com utilização de grande quantidade de imagens Landsat TM e climáticas (claras e com neblina) em áreas muito úmidas da China comprovou que estes algoritmos de remoção têm bom desempenho no processamento das imagens contaminadas por neblina. O resultado da aplicação do filtro VCP parece ser mais similar às imagens de referência quando comparado ao filtro HF. Além disso, a imagem Landsat com a remoção de neblina por VCP pode efetivamente melhorar a acurácia da classificação em comparação àquelas sem a remoção da neblina, especialmente em áreas contaminadas por nuvens. Palavras-chave:Contaminação por neblina; HF; VCP; Acurácia da classificação.

A System Dynamic Model to Estimate Leachate and Biogas Production in MSW Irregular Disposal Areas Aided by Digital Terrain Model

“…The Photoscan software optimizes the camera parameters and points feature automatically. The distance between the two coordinate (GPS and total station) calculated based on the measured coordinates was evaluated by means of the Root Mean Square Error (RMSE) as to accuracy measure, using 13 checkpoints (GCP) [65].…”

Extracting Khmer Rouge Irrigation Networks from Pre-Landsat 4 Satellite Imagery Using Vegetation Indices

“…An image convolution filter might improve results. Several cloud removal processes have been attempted but many require multiple images of the same area or the use of the blue spectral band, which is unavailable in early Landsat imagery [47]. We have tried a partial homomorphic filter process with limited success [48].…”

Comparative Analysis of Biogas Generation for the Solid Waste Sector Using Unmanned Aerial Vehicle With the Brazilian Model of Greenhouse Gas Emissions in Areas With No Operational History

“…Bulletin of Geodetic Sciences, 24(1): 98-124, Jan-Mar, 2018 It was also observed the evolution in the use of satellite images and remote sensing in different platforms, in addition to new technologies for determination of spatial characteristics that provide more accurate analyzes for different areas of knowledge as presented in Table 1. Sothe et al (2017) development of algorithms for improved land cover classification using satellite images Xiao et al (2017) photogrammetric observations for the aerial triangulation process (LIDAR 1 system) Debiasi and Mitishita (2013) comparing the use of different variables and classification algorithms for mapping (remote sensing) and monitoring of coffee growing areas. Souza et al (2016) Evaluation of characteristics of submetric images obtained with UAV for analysis of anthropic actions in landscapes.…”