Landscape Image Composition Analysis Based on Image Processing and Curve Fitting

• Published in: 2013 International Conference on Computational and Information Sciences pp. 5 - 8
• Main Authors: Hengxiang Yang, Xiujin Wang, Lifang Bai
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2013
• Subjects: Algorithm design and analysis; Clustering algorithms; Curve fitting; Educational institutions; image composition; Image edge detection; Image segmentation; normal distribution; Object segmentation; weight assignment
• DOI: 10.1109/ICCIS.2013.10

The composition of the landscape image can convey the emotion of cameramen. Proposed in this article is one novel technique for landscape image composition analysis, which could lay the foundation for the further analysis of image emotion. This method is mainly aimed at analyzing the composition of landscape image with no foregrounds compos. Firstly, the image segmentation method based on edge detection and k-means clustering algorithm is used to extract the contour of each mountain. Secondly, curve fitting method is used to determine the curve type of each mountain in the landscape image. Finally, the normal distribution model is constructed to assign a weight to each pixel of the image from its location information, We can determine the composition of the whole image based on total weight comparison of each curve type which combined the contour information and location information. In this paper, we differentiated the curve from five types: Level-style, Inclined-style, Vertical-style, C-style and W-style. Corresponding, the composition of the whole image is divided into these five types. As a result, we can determine the composition type of landscape image. Experimental results demonstrate that this method is effective and meaningful in practice.