PCA BASED FACE SKETCH SYNTHESIS USING EIGEN TRANSFORMATION

A face recognition system is based on face sketches, shape and texture information in a facial image. Therefore, automatically searching through a photo database using a sketch drawing is very useful. It will not only help the theft issues to locate a group of potential suspects, but may also help the witness and modify the sketch drawing of the suspect interaction based similar retrieved image. The proposed system contains two elements: Pseudo-sketch synthesis and Sketch recognition. The pseudo-sketch generation method is based on local linear preserving of geometry between technical image and sketch images, which is inspired by the locally linear embedding techniques. The new approach of the Facial image can be reconstructed from Eigen faces in the PCA representation. Since Eigen face is computed from the training set, the reconstructed facial image can also be expressed as the linear combination of training samples. The proposed the Eigen transformation algorithm for Principal component analysis (PCA). This method provides a powerful tool for data analysis and pattern recognition which is often used in signal and image processing as a technique for data compression, data dimension reduction or their de correlation as well, multivariate analysis or neural networks .The high frequency information or intensity can be obtained be test patch and the candidate image patches. This technique reduces the transformation error in recognition.