State of the Art of Landsat Classification Accuracy Assessment

1981

A working conference was held in Sioux Falls, South Dakota November 12-14, 1980 dealing with Landsat classification Accuracy Assessment Procedures. Thirteen formal presentations were made on three general topics: (1) sampling procedures, (2) statistical analysis techniques, and (3) examples of projects which included accuracy assessment and the associated costs, logistical problems and value of the accuracy data to the remote sensing specialist and the resource manager. Nearly twenty conference attendees participated in two discussion sessions addressing various issues associated with accuracy assessment. This paper presents an account of the accomplishments of the conference.

International Journal of Remote Sensing, 1998

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction, redistribution , reselling , loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., 2015

In compliance with the Quality Assurance Framework for Earth Observation (QA4EO) guidelines, the goal of this paper is to provide a theoretical comparison and an experimental quality assessment of two operational (ready-for-use) expert systems (prior knowledge-based nonadaptive decision trees) for automatic near real-time preattentional classification and segmentation of spaceborne/airborne multispectral (MS) images: the Satellite Image Automatic Mapper (SIAM) software product and the Spectral Classification of surface reflectance signatures (SPECL) secondary product of the Atmospheric/Topographic Correction (ATCOR) commercial software toolbox. For the sake of simplicity, this paper is split into two: Part 1—Theory, presented herein, and Part 2—Experimental results, already published elsewhere. The main theoretical contribution of the present Part 1 is threefold. First, it provides the published Part 2 with an interdisciplinary terminology and a theoretical background encompassing multiple disciplines, such as philosophical hermeneutics, machine learning, artificial intelligence, computer vision, human vision, and remote sensing (RS). Second, it highlights the several degrees of novelty of the ATCOR-SPECL and SIAM deductive preliminary classifiers (preclassifiers) at the four levels of abstraction of an information processing system, namely, system design, knowledge/information representation, algorithms, and implementation. Third, the present Part 1 requires the experimental Part 2 to collect a minimum set of complementary statistically independent metrological/statistically-based quality indicators (QIs) of operativeness (QIOs), in compliance with the QA4EO guidelines and the principles of statistics. In particular, sample QIs are required to be: 1) statistically significant, i.e., provided with a degree of uncertainty in measurement; and 2) statistically valid (consistent), i.e., representative of the entire population being sampled, which requires the implementation of a probability sampling protocol. Largely overlooked by the RS community, these sample QI requirements are almost never satisfied in the RS common practice. As a consequence, to date, QIOs of existing RS image understanding systems (RS-IUSs), including thematic map accuracy, remain largely unknown in statistical terms. The conclusion of the present Part 1 is that the proposed comparison of the two alternative ATCOR-SPECL and SIAM prior knowledge-based preclassifiers in operating mode, accomplished in the Part 2, can be considered appropriate, well-timed, and of potential interest to a large portion of the RS readership.

2014

Ideally, manufacturer specifications provide performance characteristics and specifications that can be used to evaluate the suitability of colorimeter and spectrometer measuring and test equipment for a given application. However, understanding specifications and using them to compare equipment from different manufacturers, the quality of products, and its adherence to specifications can be a perplexing task. This primarily results from inconsistent terminology, units, and methods used to develop and report equipment performance specifications. This paper discusses the continuation of work that was carried on since Hugh Fairman’s ISCC presentation in October 2012 and the ASTM adoption of the standard test method referenced herein in 2013. We review how to determine if manufacturer specifications are adequate for the intended purpose, and how to interpret and assess colorimeter and spectrometer performance and reliability. Recommended practices are presented and an illustrative exam...

… Engineering and Remote …, 1983

Remote measurement of surface temperature (Ts) allows assessing surface energy balance. However, measured radiation includes not only the radiation emitted by the surface but also the radiation emitted by the atmosphere. The signal from the surface is also attenuated by the transfer through the atmosphere. Correction of these atmospheric effects requires information on the atmospheric profiles in temperature (Ta) and vapor pressure (ea) along the atmospheric path of the thermal infrared measurement. The effect of surface emissivity must also be accounted, since it directly affects the level of emitted radiation at a given temperature. Poor knowledge in either surface emissivity or atmospheric and reflection effects results in error in the determination of Ts from remote sensing measurement. The objective of this study is to analyze the impact of using non-coincident radiosoundings to correct the Ts retrievals for atmospheric effects. We considered 27 Landsat-7 ETM+ images acquired from 2007 to 2010 over the lower Rhône Valley in France. Emissivities were estimated by considering the analysis of the NDVI -emissivity relationship together with in-situ measurements of soil and vegetation canopy emissivities performed in our study area. Ts derived from Landsat-7 were evaluated with their comparison with ground

2000

The U.S. Geological Survey, in cooperation with other gov- ernment and private organizations, is producing a conter- minous,U.S. land-cover map,using Landsat Thematic Mapper 30-meter data for the Federal regions designated by the , Environmental Protection Agency. Accuracy assessment is to be conducted,for each Federal region to estimate overall and class-specific accuracies. In Region 2, consisting of New York and New Jersey, the accuracy assessment was completed for 15 land-cover and land-use classes, using interpreted 1 :40,000- scale aerial photographs,as reference data. The methodology used for Region 2 features a two-stage, geographically stratified approach, with a general sample of all classes (1,033 sample sites), and a separate sample for rare classes (294 sample sites). A confidence index was recorded for each land-cover interpretation on the 1 :40,000-scale aerial photography The estimated overall accuracy for Region 2 was 63 percent [standard error 1.4 percent) using al...

IGARSS 2001. Scanning the Present and Resolving the Future. Proceedings. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No.01CH37217), 2001

Accuracy Assessment is one of the most important considerations in the evaluation of remotely sensed imagery. Too often, it is not done when imagery is produced. The accuracy of an image is effected by many variables, including the spatial and spectral resolution of the hyperspectral sensor, processing statistics used, types of classifications chosen, limits of detection of different surface materials, suitability of reference spectra used for image analysis training, the type and amount of ground truth data acquisition, and type of atmospheric correction algorithm applied to the imagery. This presentation will discuss selected examples generated from work performed under the NASA EOCAP (Earth Observations Commercial Applications Program) project NAS 13-99004. The first example is from the Ray copper Mine in Arizona, USA. It demonstrates the affects of spectral library references vs insitu ground truth, and different processing techniques on the identification and distribution of a target mineral, jarosite, in an image. The second example shows how the choice of processing cutoffs can change the distribution of a target mineral, alunite, in the image. The third example evaluates old and new atmospheric correction algorithms.

5 The family of Kappa indices of agreement claim to compare a map's observed classification accuracy relative to the expected accuracy of baseline maps that can have two types of randomness: (1) random distribution of the quantity of each category and (2) random spatial allocation of the categories. Use of the Kappa indices has become part of the culture in remote sensing and other fields. This article exam-10 ines five different Kappa indices, some of which were derived by the first author in 2000. We expose the indices' properties mathematically and illustrate their limitations graphically, with emphasis on Kappa's use of randomness as a baseline, and the often-ignored conversion from an observed sample matrix to the estimated population matrix. This article concludes that these Kappa indices are useless, mis-15 leading and/or flawed for the practical applications in remote sensing that we have seen. After more than a decade of working with these indices, we recommend that the profession abandon the use of Kappa indices for purposes of accuracy assessment and map comparison, and instead summarize the cross-tabulation matrix with two much simpler summary parameters: quantity disagreement and alloca-20 tion disagreement. This article shows how to compute these two parameters using examples taken from peer-reviewed literature.

1983

Research performed in the accuracy assessment of remotely sensed data is updated and reviewed. The use of discrete multivariate analysis techniques for the assessment of error matrices, the use of computer simulation for assessing various sampling strategies, and ...