The Impact of Positional Errors on Soft Classification Accuracy Assessment: A Simulation Analysis

Abstract

Abstract:Validating or accessing the accuracy of soft classification maps has rapidly developed over the past few years. This assessment employs a soft error matrix as generalized from the traditional, hard classification error matrix. However, the impact of positional error on the soft classification is uncertain and whether the well-accepted half-pixel registration accuracy is suitable for the soft classification accuracy assessment is unknown. In this paper, a simulation analysis was conducted to examine the influence of positional error on the overall accuracy (OA) and kappa in soft classification accuracy assessment under different landscape conditions (i.e., spatial characteristics and spatial resolutions). Results showed that with positional error ranging from 0 to 3 soft pixels, the OA-error varied from 0 to 44.6 percent while the kappa-error varied from 0 to 93.7 percent. Landscape conditions with smaller mean patch size (MPS) and greater fragmentation produced greater positional error impact on the accuracy measures at spatial resolutions of 1 and 2 unit distances. However, this trend did not hold for spatial resolutions of 5 and 10 unit distances. A half of a pixel was not sufficient to keep the overall accuracy error and kappa error under 10 percent. The results indicate that for soft classification accuracy assessment the requirement for registration accuracy is higher and depends greatly on the landscape characteristics. There is a great need to consider positional error for validating soft classification maps of different spatial resolutions. OPEN ACCESSRemote Sens. 2015, 7 580

Árvore De Decisão E Análise Baseada Em Objetos Na Classificação De Imagens Com Resolução Espacial Submétrica Adquiridas Por Vant

“…Para cada construção do modelo é dada mais atenção às regras de classificação com maiores taxas de erros, tentando melhorá-las no próximo ensaio (Quinlan, 2013 Por meio das amostras de validação formadas com base nos 500 pontos aleatórios, foi calculado para cada classificação o índice kappa, para isso se utilizou a biblioteca StatsModels. O índice kappa é uma técnica discreta multivariada utilizada na determinação da acurácia de mapas temáticos (Gu et al, 2015). Também foi utilizada uma classificação qualitativa dos valores de kappa, conforme o trabalho de Landis e Koch (1977).…”

“…Considerando a segmentação, os parâmetros do algoritmo C5.0 e as classificações por SVM, o modelo de classificação com maior valor de kappa de cada algoritmo foi utilizada para avaliar o erro de classificação em relação às classes de cobertura da terra. Esse erro de classificação por classe foi calculado por meio da Acurácia do Produtor (AP) (Gu et al, 2015).…”

Changes in crop type distribution in Zhangye City of the Heihe River Basin, China

“…In this study, based on the land cover data in June of 2012(DMCHRPC, 2012Zhong et al, 2014), 50 verification samples each for corn, wheat, barley, and oilseed rape were selected as the real reference source using a stratified pure pixel sampling method. An accuracy assessment was performed based on the error matrix method (Congalton, 1991;Gu et al, 2015); and the mapping accuracy, user accuracy, overall accuracy, and Kappa coefficient of the crop classification results in Zhangye City of 2012 were also all calculated.…”

“…High temporal resolution remote sensing images are most suitable for the classification of simple crop sowing patterns in a region with a large field size (Sun, Xu, Lin, Zhang, & Mei, 2012;Zhang, Lei, Wang, Li, & Zhao, 2011). However, in those areas with fragmented cultivated land and rather complex crop planting patterns, the classification accuracy is usually unsatisfactorydbecause of the problem of mixed pixelsdwhen MODIS or other high temporal resolution remote sensing images with low spatial resolution serve as the data source (Gu, Congalton, & Pan, 2015;Jain, Mondal, DeFries, Small, & Galford, 2013;Wu & Li, 2012). Compared with these lower spatial resolution images, Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETMþ) images offer higher spatial resolution at a proper temporal resolution (30 m and 16 days, respectively).…”

An Upscaling Algorithm to Obtain the Representative Ground Truth of LAI Time Series in Heterogeneous Land Surface

“…ASTER 2B05V data are atmospheric corrected surface reflectance with three bands (0.52-0.86 μm) and 15 m spatial resolution in UTM (WGS84) projection. All images were geometrically corrected to the standard reference image with 0.5 pixel accuracy, which was commonly accepted based on the findings of Gu et al [44]. Several LAI products including two MODIS collection 5 (C5), GLASS v3.0, and GEOV1 (Table 2), were collected and used for comparison with the ground truth of LAI created by the proposed method detailed in Section 3.…”