In the case of an intelligent biometric system, identifying facial features is a challenging task. Facial features such as eyes, mouth, lips are the critical factor for expressing human emotions. Face and facial feature detection can be implemented automatically with the help of the computer, but it is a difficult job. In this paper, we proposed a new framework for fast and efficient detection of face and facial features such as eyes, nose, mouth, and lips from color images. Here face image is given as input and facial features like eyes, mouth and lips are given as output. The performance of our algorithm is developed from skin color segmentation and is based on three stages, face detection, region localization and facial feature detection. Our proposed algorithm is modified from RGB and HSV color space, which offers better performance. Our experimental result shows that the proposed algorithm is better at detecting facial features than Viola and Jones based on the statistical method. Reduces the problem of image position, expression and lighting variation. The average accuracy of our algorithm is 97.69%, and it is easy to extract facial features from color images. Face detection with keywords; regional localization; skin color segmentation; eye detection; lip detection.I. INTRODUCTIONFace and facial feature extraction is one of the most challenging problems in disciplines such as image processing, pattern recognition, and computer vision. Due to different poses, facial expressions, orientation, lighting conditions, colors of the images. With the improvement of information technology, facial feature identification has found wide use in the application, such as personal identification, video surveillance, AVSR (audio-visual speech recognition), witness face... half of paper... C , LHS, RGB (red, green, blue) and modified RGB etc. A pixel-based skin color detection method is introduced in this paper because it is faster than other methods. It can classify each pixel as skinned or non-skinned by its neighbors. In this method we first detect the face from the image. The YCbCr [10] [16] color space is used for skin color segmentation of facial features. The eye and mouth regions are extracted from the face image. The facial region around the regions is eliminated by applying the threshold value. After facial region deletion, facial features are detected from the whole image by color segmentation, but they have small unwanted pixels, which are reduced by image erosion and dilation through the morphological process. Now the image is converted to binary image and connects existing components to detect facial features such as eyes, mouth and lips.