A classification method of hyperspectral reflectance images named CHRIPS (Classification of Hyperspectral Reflectance Images with Physical and Statistical criteria) is presented. This method aims at classifying each pixel from a given set of thirteen classes: unidentified dark surface, water, plastic matter, carbonate, clay, vegetation (dark green, dense green, sparse green, stressed), house roof/tile, asphalt, vehicle/paint/metal surface and non-carbonated gravel. Each class is characterized by physical criteria (detection of specific absorptions or shape features) or statistical criteria (use of dedicated spectral indices) over spectral reflectance. CHRIPS input is a hyperspectral reflectance image covering the spectral range [400–2500 nm]. The presented method has four advantages, namely: (i) is robust in transfer, class identification is based on criteria that are not very sensitive to sensor type; (ii) does not require training, criteria are pre-defined; (iii) includes a reject class, this class reduces misclassifications; (iv) high precision and recall, F ${}_1$ score is generally above 0.9 in our test. As the number of classes is limited, CHRIPS could be used in combination with other classification algorithms able to process the reject class in order to decrease the number of unclassified pixels.

The CHIWAWA (Cloud detection in Hyperspectral Images With Atmospheric column WAter vapor) cloud detection method is presented. It aims to detect thick and thin clouds in a single hyperspectral image covering the reflective spectral range [400-2500nm] in an unsupervised way. Its main originality is to combine the complementary information available in the ground reflectance image and the atmospheric column water vapor map (WV map). The method uses criteria based on reflectance to compute a mask of potential cloud pixels and refine it by exploiting the WV map. The use of the WV map is central to the method, as it has the greatest discriminating power. It therefore plays a role similar to that of thermal bands, which are very useful for cloud detection, but are absent in the reflective spectral range. Two specific processings are also proposed to detect snow and ice pixels and to reduce the altitude dependency of the WV map in mountain images. CHIWAWA performance is assessed on several hyperspectral images acquired with PRISMA and AVIRIS-NG sensors. Overall accuracy ranges in 97–100% for both sensors. When cloud cover is above 10%, F ₁ score ranges in 94–100% for both sensors. For lower cloud covers, F ₁ score ranges in 83–94% for PRISMA images and in 84–91% for AVIRIS-NG images. CHIWAWA may encounter difficulties in detecting thin clouds located over urban areas or when the contrast between the cloud and its surroundings is low in the WV map.