KEYWORDS: RGB color model, Image retrieval, Curium, Object recognition, Mathematical modeling, Reflectivity, Data modeling, Light scattering, Sun, Information theory
Color, which is effective information for computer vision, is vulnerable to variations in illumination. It is necessary to achieve illumination invariance for the description of an image’s color information. We propose two new color spaces that are robust to illumination changes: the color information color space based on the diagonal-offset model and the edge color space, the two channels of which are derived from the spherical color space and the linear transformation between the dark channel and the bright channel. Then, we propose a color moment invariant descriptor according to Hu moment invariants in the two new color spaces. We test the performance of the color descriptors with illumination variations in the context of object recognition and image retrieval. Experiments show that our descriptor has satisfactory stability for light variations.
Illumination estimation algorithms are aimed to estimate the RGB of scene illumination color when the image was taken, which is a significant way to achieve color constancy. They can be divided into three categories: pixel-based algorithms, learning-based algorithms and combination algorithms. Compared with other two kinds of illumination estimation algorithms, pixel-based algorithms are relatively poorly performing. In this paper, we add a L0-norm smoothing preprocessing to pixel-based algorithms to improve the performance. The L0-norm smoothing can suppress insignificant details and maintain major edges of an image. Experimental results show that our optimization approach is effective to enhance the performance of pixel-based algorithms.
Illumination estimation is an important part of color constancy. It is known that estimating the scene illumination color is an ill-posed problem, and no method can be considered as universal. To select an optimal method for a particular image or scene, some combination methods based on characteristic similarity were proposed. These methods found that the images having similar characteristics can use the same method as an optimal technique. Although the combination methods based on image characteristics have worked well in some scenes, the accuracy of these methods is limited to the accuracy of the set of the unitary method. However, we assume that images with similar characteristics have similar scene illumination color and propose an illumination estimation method which is based on image characteristics. According to the characteristics of each image, we search K images whose characteristics are similar to the image from the image dataset, and the standard illumination color of the K selected images is known. Then, we use the weighted average method to estimate the new image’s illumination color by combining the standard illumination color of the K selected images. The experimental results show that the proposed method outperforms other state-of-the-art methods.
Trunk ranging is an essential function for autonomous forestry robots. Traditional trunk ranging systems based on personal computers are not convenient in practical application. This paper examines the implementation of a trunk ranging system based on the binocular vision theory via TI’s DaVinc DM37x system. The system is smaller and more reliable than that implemented using a personal computer. It calculates the three-dimensional information from the images acquired by binocular cameras, producing the targeting and ranging results. The experimental results show that the measurement error is small and the system design is feasible for autonomous forestry robots.
For the intelligent pruning machine, a machine vision system is pre-requisite. Standing tree image segmentation is a key
step for the machine vision system. An efficient scheme for tree image segmentation was proposed according to the need
of the machine vision system of the intelligent pruning machine. The scheme is a level set method based on particle
swarm optimization. According to principal of the level set method, the image segmentation is formulated as one of
optimization problems. The energy function is taken as the segmentation quality criteria, which consists of an internal
energy term that penalizes the deviation of the level set function from a signed distance function, and an external energy
term that drives the motion of the zero level set toward the desired image feature, such as object boundaries. In this
paper, the method used particle swarm optimization to solve the optimization problems that is different from the ordinary
level set method that uses the partial differential equation method in some literatures. In experiments, tree images with
different background are selected to test the efficiency of the scheme that presented in this paper. In order to test the
antimonies performance of the scheme that presented in this paper, a tree image added Gaussian white noise is selected.
From the results of the tree image segmentation, the scheme that presented in this paper is more efficiently. The
experimental results demonstrate the scheme is more effective and time-saving than the ordinary level set method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.