An edge detection method using 2-D autoregressive lattice prediction filters for remotely sensed images

Edges characterize boundaries and are therefore a problem of fundamental importance in image processing. Edge detecting an image significantly reduces the amount of data and filters out useless information, while preserving the important structural properties in an image. Edge detection is useful for segmentation, registration, and identification of objects in remote sensing images. Two dimensional lattice filters have been shown to be useful in many applications such as multidimensional spectral estimation, image data compression, high-resolution radar imaging, and removal of correlated clutter to enhance the detection ability of small objects in images. In this work, lattice filters are used for detecting the edges in remote sensing images. Lattice filter can be used to predict the correlated parts in an image and the resulting error (the output of the filter) will be edges. Edge detection results have been compared with other conventional edge detection methods as well as wavelet based methods. Results show that the proposed method is a good candidate for edge detection problem in remotely sensed images