Liu Hongzhe

GrabCut image segmentation algorithm based on structure tensor

Abstract This paper attempts to present an interactive color natural images segmentation method. This method extracts the feature of images by using the nonlinear compact structure tensor (NCST) and then uses GrabCut method to obtain the segmentation. This method not only realizes the non-parametric fusion of texture information and color information, but also improves the efficiency of the calculation. Then, the improved GrabCut algorithm is used to evaluate the foreground target segmentation. In order to calculate the simplicity and efficiency, this paper also extends the Gaussian mixture model (GMM) constructed base on the GrabCut to the tensor space, and uses the Kullback-Leibler (KL) divergence instead of the usual Riemannian geometry. Lastly, an iteration convergence criterion is proposed to reduce the time of the iteration of GrabCut algorithm dramatically with satisfied segmentation accuracy. After conducting a large number of experiments on synthetic texture images and natural images, the results demonstrate that this method has a more accurate segmentation effect.

Research on quantifying color of HSV in extraction of middle-level semantic objects of traditional Chinese painting images

Traditional Chinese painting is a unique form of art and highly regarded for its theory, expression and techniques throughout the world. The scripts part and the seals part of the traditional Chinese painting images are integral parts of traditional Chinese paintings and have high semantics. The automatic extraction of these two parts becomes a key problem. The seals part and scripts part are mostly in black ink and red color respectively. To extract them, the black or red regions should be first extracted. In the paper, we focus on how to quantify color of HSV to get better extracted results according to the characteristics of traditional Chinese painting images. A color quantifying model is given first. Then the results under different quantifying value are analyzed in details. The extracted results are satisfactory. The work of the paper is very meaning to further extract the scripts part and the seals part of traditional Chinese painting images accurately.

Towards a Framework for agent-enabled Semantic Integration in the design of Distributed Digital Museum

This paper presents a framework for agent-enabled semantic integration in the design of distributed digital museum. In this framework, Web Services provides remote information process services which can be discovered and invoked by middle-layer agent to satisfy multiple needs. Domain Ontology is used to characterize and model the semantics that are embedded in the digital museum. Concept-mapping Agent is deployed to set up the semantic relationships between Domain Ontology and relational databases and feedback to middle-layer agent, in doing so, we can get directly benefit from middle-layer agent. During the user request answer process, middle-layer agent can dynamically and quickly discovered the services, executed the remote calling of services, combined the results and finally feedback to user. Furthermore, the paper gave out a Digital Museum Information Retrieval System (DMIRS) , which is designed, based on this framework and demonstrated to be stable and efficient.

Lateral control of autonomous vehicles based on learning driver behavior via cloud model

Abstract In order to achieve the lateral control of the intelligent vehicle, use the bi-cognitive model based on cloud model and cloud reasoning, solve the decision problem of the qualitative and quantitative of the lateral control of the intelligent vehicle. Obtaining a number of experiment data by driving a vehicle, classify the data according to the concept of data and fix the input and output variables of the cloud controller, design the control rules of the cloud controller of intelligent vehicle, and clouded and fix the parameter of cloud controller: expectation, entropy and hyper entropy. In order to verify the effectiveness of the cloud controller, joint simulation platform based on Matlab/Simulink/CarSim is established. Experimental analysis shows that: drivers lateral controller based on cloud model is able to achieve tracking of the desired angle, and achieve good control effect, it also verifies that a series of mental activities such as feeling, cognition, calculation, decision and so on are fuzzy and uncertain.

The design and implementation of ontology based image retrieval for Xu Beihong Museums

Ontology is formal shared conceptualization of domains of interest providing common vocabularies and used more and more for information retrieval system. In the paper, ontology is used in XuBeihong Museums retrieval system. A framework based on ontology is proposed. How to create knowledge base and supplement it, how to retrieve are dissertated. In addition, the virtues of retrieval based on ontology are compared with retrieval based on keywords.