Abstract We have designed a facial image processing and retrieval model with the help of the content. We have deployed the SIFT descriptor for detecting key points in the target face. Also, we have applied Naïve Bayes classifier to find the distance between two faces and by reducing key points. We have tested the system and found the performance.... We have designed a facial image processing and retrieval model with the help of the content. We have deployed the SIFT descriptor for detecting key points in the target face. Also, we have applied Naïve Bayes classifier to find the distance between two faces and by reducing key points. We have tested the system and found the performance. Read More Read Less

Abstract In the CT images, during compression, the loss of quality is reported and hence research has been conducted to offset this limitation. We in this paper, have presented a lossless compression system and to ensure complexity. The Hierarchical Adaptive Karhunenloeve transform method is used which ensured good energy for the decorrelation process. It is found to have better compression rate... In the CT images, during compression, the loss of quality is reported and hence research has been conducted to offset this limitation. We in this paper, have presented a lossless compression system and to ensure complexity. The Hierarchical Adaptive Karhunenloeve transform method is used which ensured good energy for the decorrelation process. It is found to have better compression rate than others. This process helped to achieve the high decorrelation for the whole population of images. The main energy in the images is found to be place over the eigen images. The new algorithm introduced helped to decrease the complexity and found to be more effective. Read More Read Less

Abstract In the JPEG compressed images and I-frames of the MPEG compressed videos, the text region extraction is challenging which can be carried out through localization. We in this paper have identified the text location with the text regions using DCT compression which is in contrast to conservative full decompression of the videos. Due to the advantage of less decoding, the... In the JPEG compressed images and I-frames of the MPEG compressed videos, the text region extraction is challenging which can be carried out through localization. We in this paper have identified the text location with the text regions using DCT compression which is in contrast to conservative full decompression of the videos. Due to the advantage of less decoding, the compression process seems to be more faster. This process ensures text detection in large sized images. While we applied the algorithm we are able to arrive at effective results. Read More Read Less