The Hyperspectral Mineralization Recognition Technology Based On Spectral Matching Study

During the last 20 years, hyperspectral remote sensing has been playing an important role in many fields. Compared to Multi-spectrum remote sensing, the characteristics of hyperspectral remote sensing data are more of channels, higher spectral resolution, narrower bandwidth and larger amount of data,all of these characteristics in the surface mineral classification and recognition has a great advantage. so, it has being widely used in geology survey field, such as mineral recognition and mineral mapping. In these applications, the spectral mapping technique is one of the key techniques.In this paper, we focus on the hyperspectral mineral recognition technique and then do research around this subject. So, the importance is the research about spectral mapping technique and a useful and effective database. At first, we use the database(PIMA) and the infrared spectrometric analyzer(PMS) to bulid a useful and effective database; then, on this basis, contrast and analyse some spectral mapping technique in common use, develope and validate a new spectral mapping technique, this new technique use all of the curve's feature and strengthens the single diagnostic feature to improve the rate of mineral recognition. We also do some experiments from mineral recognition and spectral classification to validate this technique, then summarize its advantages and disadvantages. Lastly, according to the database and these mapping techniques,the author codes the algorithms and carry out it on database data to experiment and approve the algorithms' veracity, it's not only for people in the laboratory to recognize mineral, but also for the people in the wild to extract the information of surface.