We considered fitting the lineaments to the set of the points extracted from the picturewhich expressed the geographical structures. In the case when a coefficient of linear equation which expresses the lineament is calculated by the method of least square for distance between thepoints and a straight line, the fittings among the linear equations of each sets of points can becompared by the averages of the residual sum of squares.
In this paper, all points are expressed on a plane coordinates and objective variable y is expressed by the function f (x) .When the coefficient of f (x) to minimize the residual sum of squares between measured value yi and f (xi) is calculated, accuracy of the function can be expressed by the correlation coefficient.
Therefore, we moved the origin of coordinates to the center of set of points and rotated the coordinates at the point to calculate the maximum correlation coefficient. Moreover, it was proved that the linear equation with the maximum correlation coefficient was equivalent to the straight line derived by the method of least square for distance between the points and a straight line.
We showed the comparison results of the accuracy of the fitting lineament on the region where gold deposit exists as an example of application of this method.
View full abstract