** Progress in Earth and Planetary Science is the official journal of the Japan Geoscience Union, published in collaboration with its 51 society members.
** Progress in Earth and Planetary Science is partly financially supported by a Grant-in-Aid for Publication of Scientific Research Results to enhance dissemination of information of scientific research.
Gallery View of PEPS Articles
Methodology
Space and planetary sciences
Segmentation of dust storm areas on Mars images using principal component analysis and neural network
Gichu R, Ogohara K
Mars, Dust storm, Segmentation, Machine learning, Principal component analysis
(Left) Example of red band (575–675 nm) image of the area around 180°E-40°N extracted from a global swath image observed by Mars Orbiter Camera onboard Mars Global Surveyor, which has been already preprocessed based on the main text. We can see a distinct dust storm at the center of the image. (Right) Probability image of dust storm created by the proposed method.
We present a method for automated segmentation of dust storm areas on Mars images observed by an orbiter. We divide them into small patches. Normal basis vectors are obtained from the many small patches by principal component analysis. We train a classifier using coefficients of these basis vectors as feature vectors. All patches in test images are categorized into one of the dust storm, cloud, and surface classes by the classifier. Each pixel may be included in several dust storm patches. The pixel is classified as a dust storm or the other classes based on the number of dust storm patches that include the pixel. We evaluate the segmentation method by the receiver operator characteristic curve and the area under the curve (AUC). AUC for dust storm is 0.947–0.978 if dust storm areas determined by our visual inspection are assumed to be ground truth. Precision, recall, and F-measure for dust storm are 0.88, 0.84, and 0.86, respectively, if we remove false negative pixels efficiently and maintain the size of true positive dust storms using two different threshold values. The tuning parameters of the classifier used in this study are determined so that the accuracy for dust storm is maximized. We can also tune the classifier for cloud segmentation by changing the parameters.
