Journal of Scientific & Industrial Research Vol. 78, April 2019, pp. 193-198 Multi-Operator Content Aware Image Retargeting on Natural Images M Abhayadev 1 * and T Santha 2 1,2 Department of Computer Science, Dr G R Damodaran College of Science, Coimbatore, Tamil Nadu – 641014, India Received 22 June 2018; revised 13 December 2018; accepted 04 March 2019 The image retargeting can be done in two methods, i.e. the traditional image retargeting methods (such as cropping and scaling) and content aware image retargeting methods. This research paper proposes a Multi-Operator Content Retargeting Image resizing Technique (CRIST) for image retargeting. The proposed retargeting operator is a combination of the visual saliency based scaling operator and the seam carving. The efficiency of the proposed research is experimented on a natural image dataset. CRIST works well on all natural images based on subjective and objective quality metrics. Keywords: Target Prediction, CRIST, Filters, DoG, Erosion, Peak-Signal-to-Noise Ratio Introduction Image media retargeting has been retained as a very important research topic in the digital image processing research. In order to protect certain visual saliency in important areas, so many resizing 1,2 methods are introduced by Zhang LumingandYan Bo , using the significance map and the importance map which are based on local low levels and high level features. The table no.‗1‘ represents different types of content aware image retargeting operators in the image resizing research. Authors introduced an efficient retargeting of shadow images 3 by using Multi-Operator improved- CRIST. Another article of the same authors also proposed a content aware image seam carving technique 4 for object resizing which could preserve the salient objects in images. Materials and methods Image retargeting is generally considered as one of the best content aware image resizing technologies in the area of digital image processing research. The proposed technique will work better on the natural images and the proposed method efficiency could be compared to the existing state-of-the-art image retargeting operators. The proposed Multi-Operator Content Retargeting Image reSizing Technique (CRIST) contains three stages. They are the following  Image object boundary identification by using enhanced high pass frequency filtering algorithm and morphological erosion structuring element.  Location based visual saliency map generation.  Proposed natural image retargeting based on hybridized Multi-Operator. Figure 1 presents the framework of the proposed technique for content aware image retargeting. This framework is explained in the following sub-sections: Image object boundary identification by using enhanced high pass frequency filtering algorithm and Morphological Erosion Structuring Element The first stage in content aware image retargeting is the process of identifying the image objects boundary. This particular method uses a high pass frequency filtering algorithm followed by morphological erosion structuring element to identify object boundary. The Difference of Gaussian (DoG) high pass filter is applied for identifying the object boundary. The high pass filter works efficiently in finding the high energy values, gradient amplitude value of pixels in input images. Image thresholding will segment the image into different parts according to the region of significance and less significance. Finally the important object boundary will be normalised by using morphological erosion structuring element. Location based visual saliency map generation The proposed visual saliency map works on the location based saliency map technique. The important image‘s object features are identified by using both top-down and bottom-up saliency computation methods. The Location Based Visual Saliency ————— *Author for Correspondence E-mail: abhayadevmalayil@gmail.com