This paper presents a color recognition algorithm for natural scene images based on HSV color space.In this algorithm,different fuzzy membership are constructed on the hue channel,saturation channel and value channel.By establishing the fuzzy rule base,the Mamdani fuzzy inference method is selected,and the color images are divided into eleven kinds according to the human psychological perception of color.This algorithm can accurately identify the color of the image with rich color distribution and partial background color close to the target color. Experiments show that the algorithm's color recognition and segmentation results are in good agreement with human subjective visual perception, which effectively improves the color recognition accuracy of the robot.
Compared with grayscale images, color images with color information have more advantages in target recognition, detection and tracking, but the color of the image can be easily affected to be distorted by light. Aiming at solving the problem of color distortion, a method based on wavelet transform is proposed to eliminate the influence of illumination. Firstly, the V-channel components in HSV space are decomposed by wavelet and the illumination component in the image scene is estimated by reconstructing the approximate coefficients of wavelet decomposition. Secondly, according to the imaging principle model, the illumination component is removed from the original image, and the reflection component that characterizes the color information is preserved. Finally, the normalized distribution function is used to normalize the reflection components, and the image color recovery is achieved in RGB space. The experimental results show that the proposed method can reduce the influence of non-uniform illumination on the image to restore the color and improve the image quality.
To identify the accuracy of the installation position of the spring hooks on both sides of the automobile seat back curtain automatically, this paper presents an automatic detection system for spring hooks on automobile seat back based on feature detection and line segment. Firstly, the backrest image of the automobile is matched by a scale-invariant feature transformation algorithm (SIFT algorithm) to classify different types of automobile seat backs. Secondly, an edge detection method is used to extract the characteristic thread to locate the middle line of the spring hook (or parallel lines) and the side line where the spring hook hooks. Finally, the possibility of the intersection of the spring middle line (or parallel lines) and side line to be inside the side line slot is calculated to determine qualification level of the automobile seats. The experimental results indicated that the system detection accuracy was up to 98.8%. The system meets the practical requirement of automobile industry with high efficiency.
In allusion to the problem that the near infrared images of early teeth disease detection is low contract, minutiae feature isn’t evident and the images have small target gray range. So the image enhancement approach is proposed, which is disposed by the high frequency filter, linear compensation and the successive mean quantization transform. First, the near infrared images of teeth are filtered by the high frequency filter, the high frequency filter can emphasize on high-frequency components and keep lost low-frequency components, then the near infrared images are linearly compensated, finally the method combined with the successive mean quantization transform to enhance the near infrared images. The experiments show that the enhancement images show high contract and detail feature is obvious. Visually it is better to observe whether the teeth are diseased. Compared with the results of histogram equalization, the proposed method can get better enhancement effect and the particular features are stand out.
Hand gesture has been considered a natural, intuitive and less intrusive way for Human-Computer Interaction (HCI). Although many algorithms for hand gesture recognition have been proposed in literature, robust algorithms have been pursued. A recognize algorithm based on the convolutional neural networks is proposed to recognize ten kinds of hand gestures, which include rotation and turnover samples acquired from different persons. When 6000 hand gesture images were used as training samples, and 1100 as testing samples, a 98% recognition rate was achieved with the convolutional neural networks, which is higher than that with some other frequently-used recognition algorithms.