ICTACT Journal on Image and Video Processing

Satish S. Hiremath ¹, K. V. Suresh ¹

Affiliations
1 Department of Electronic and Communication Engineering, Siddaganga Institute of Technology, IN

Abstract

An important task in content based video indexing is to extract text information from videos. The challenges involved in text extraction and recognition are variation of illumination on each video frame with text, the text present on the complex background and different font size of the text. Using various image processing algorithms like morphological operations, blob detection and histogram of oriented gradients the character recognition of video subtitles is implemented. Segmentation, feature extraction and classification are the major steps of character recognition. Several experimental results are shown to demonstrate the performance of the proposed algorithm.

Keywords

Character Recognition, Classification, Complex Background, Feature Extraction, Segmentation.

Full Text

     

S. Renukalatha ¹, K. V. Suresh ²

Affiliations
1 Department of Computer Science and Engineering, Sri Siddhartha Institute of Technology, IN
2 Department of Electronics and Communication Engineering, Siddaganga Institute of Technology, IN

Abstract

Accurate segmentation of medical images is pivotal in medical image analysis as it favors the detection and quantification of abnormalities present in human anatomical structures. Since medical images are complex and sometimes noisy, effective extraction of the regions of abnormalities is a tedious process. Many semi-automatic segmentation algorithms with appreciable segmentation accuracy do exist in literature. However, these techniques are iterative, computationally expensive, involve human intervention demanding initial parameter settings and moreover, each one of them is specific to a particular modality. In addition, presence of noise further degrades the quality of the processed image. There is no general algorithm to extract the key regions from all types of noisy medical images. This paper proposes an automatic Region of Interest (ROI) extraction algorithm to detect the important regions in noisy medical images of different modalities using statistical moments. The proposed approach estimates an optimal threshold value automatically using statistical moments through histogram decomposition technique. Initially, the medical image database is preprocessed followed by ROI extraction and the performance of the proposed approach is compared with other techniques to verify its robustness.

Keywords

Histogram, Optimal Threshold, Moments, ROI, Medical Imaging Modalities.

Full Text