著者:六川修一
著者:六川修一
原画像(左画像)は ©METI and JAXA[2007] Distributed by ERSDAC 著者:六川修一
著者:六川修一
著者:村松加奈子
著者:村松加奈子
著者:村松加奈子
著者:村松加奈子
著者:村松加奈子
出典:山形与志樹、安岡義文、宮崎忠国、小熊宏之、森山隆、中谷幸広、ランドサットTM画像を用いた火災による釧路湿原の焼失面積の推定、日本リモートセンシング学会誌, Vol.12, No.4, pp71-76, 1992 著者:村松加奈子
著者:村松加奈子
Chapter 12 Image Enhancement and Feature Extraction: Spatial and Temporal Information
Differential processing (a) Original data (b) First derivative (c) Second derivative
Filter processing (a) Filter operator (b) An image to be processed (c) Result of the filter processing
Sobel filter (a) X direction operator (b) Y direction operator
Laplace operator
Original image (radar), PALSAR image around Mt Original image (radar), PALSAR image around Mt. Wanitsuka in Myazaki prefecture ©METI and JAXA (2007), source: JSS
Result of Sobel filter in the horizontal (X) direction
Combined image of Laplace operation and the original image (image sharpening)
A sample image window (Figure 12.8)
Relationships between adjacent pixels i: a value of a reference pixel j: a value of a neighbor pixel (right of the reference pixel)
The gray level co-occurrence matrix for the image window, Figure 12.8
A procedure for the extraction of time series variation
Geometric registration for images in a time series sequence
Cloud removal
Demonstration of change areas based on color composite (Yellow: areas without change)
Extraction of variation by differences between data from two dates
Frequency distribution for the values of NDVI differences
Extraction of variations based on the analysis of principal components Occurrence of fire In Kushiro wetland in November 2, 1991
Principal component analysis using TM data of 3 bands(R, NIR, SWIR) for two image dates