Forest Cover and Land Use/Land Cover Mapping
of Mazandran Province, using Remote Sensing
Satellite Tm Data, to Prepare Atlas of Study Area
Image Processing and Classification
Digital image processing is the numerical manipulayion of digital image and includes
preprocessing, enhancement and classification, preprocessing refer to the internal processing of the raw data to calibrate the image radiometry, correct geometric distortions and remove noise the nature of the particular preprocessing required obviously depends strongly on the sensors characteristics, because the preprocessing is designed to remove any sensor the corrected image are then submitted to enhancement produces a new enhanced image that is displayed on a (CRT) for classification, this enhanced image may be easier to classify than the original in different way.
Supervised Classification
Supervised classification proceduses are the eassential analytical from the extraction of quantitative information from remotely sensed image data to decide the set of ground cover type in to which the image is to be segmented these are the information classes just like urban, area water
bodies crop land. Range land, than choose reprecentive or prototypc pixels from each of the desried set of classes, these are pixel as a train data, training sets for each class canbe estabilished using the training dat ato estimate the parameters of the particular classification algorithm to be used. These parameters for a given class make signature of that class so that these signature apply the trained classify every pixel in the image in to one of the desired ground cover type (information classes) here the whole image segment of interest is typically classified, then these classification are overlay with interpretation base map joint together. Show fig.
image of study area 1993
image of study area 1994
Scatter Plot of study area 1993
image of study area 1993
Landuse mao legend
Conclusion:
This project has been done by using Landsat thematic Mapper ™ data for mazandaran Province and Could effectively and quickly be used For monitoring of landsat/ land cover.
The agriculature units were delineated based on color theme differences on the TM images through visual interpretation and were corrected by combination of digital analysis of TM data and the data obtained from ground truth.
