Survey on Classification and Segmentation for HDR Satellite Imagery

Electronics and Telecommunications

Now a day's the most interesting research area in field of Digital Image Processing is Remote sensing. Currently Remote sensing (RS) is being widely used technology which plays most important role in Image analysis. Remote Sensing (RS) generally refers to the use of satellite based sensor technologies respectively. However, the luminance of natural scene present over a large dynamic range. Thus, with a regular camera it becomes complicated to capture the details of an entire scene with good clarity in a single shot. To overcome this problem HDR (High Dynamic Range) imaging has been used. TMO (Tone Mapping Operator) is used to convert HDR image into LDR image which serves good quality of an image, so that we can visualize it over LDR display. Currently in the field of Image processing an image with big data becomes quite difficult to understand. To overcome this analysis problem there is need of image segmentation. Image segmentation is used to compress the data which will helpful to understand. Satellite imagery is one of the wide areas in segmentation and classification. There are various image segmentation techniques are there like edge, threshold, region, clustering and neural network segmentation method. This work focuses image segmentation towards ANN (Artificial Neural Network) because it has some advantages as compare to rest of methods. Image classification techniques like supervised and unsupervised learning discussed in this paper.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.