Hyperspectral Remote Sensing of Forests: Technological advancements, Opportunities and Challenges

In real world what we are able to see is just because of light or energy reflected or emitted from the viewing object is falling upon retina of human eye. The variations in intensity of light reflected back from any object in different wavelengths are sensed and provide ability of discriminating different objects having similar size and shape. In the same way, in spectroscopy we sense the reflected light through artificial sensors and record as image (in airborne and satellite spectroscopy) or as spectrum (in field spectroscopy). In remote sensing discrimination of different object mainly depends on difference in reflection of energy in different wavelength region of light. Considering this behaviour of light, in hyperspectral remote sensing the reflected light coming from object is split into multiple continuous and small-small wavelength bands and are sensed in each wave band separately. Therefore we are having reflection response of object in multiple and narrow wavelength regions, which can be used in discrimination of different objects that are not separable in multispectral remote sensing due to less number of broad range wave bands. Collection of data is one aspect of the technology but as soon as these data are collected, a question arises how to and where to use this data? To answer where to use, a list of applications like discrimination, mapping and monitoring of different features and process of landforms in ecosystem have been reported, and forestry is one of them. And question of how to use these data in each applications involve converting the raw data into useful information using a multistep process of atmospheric, radiometric and geometric correction, removal of bad data and data redundancy, transformation and extraction of most useful data, data segmentation and extraction of useful information. For this purpose variety of data processing techniques, algorithms, concepts and schemes have been reported from time to time. In this review article we have summarized the available technical developments in hyperspectral remote sensing during the last three decades and tried to discuss the opportunities and challenges in hyperspectral remote sensing applications in the forestry sector.