Earticle

PGQ, a novel hybrid computational intelligence approach, in which Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and quantum computation are integrated, is proposed to solve the Vehicle Routing Problem (VRP). In PSO, a quantum approach called QUP is proposed to update the particles. GA operators are employed to improve population quality. The simulation results indicate that the PGQ algorithm is very effective and is better than simple PSO and GA as well as PSO and GA mixed algorithm.

This paper presents a hierarchical deformable model for robust human face detection, especially with occlusions and under low resolution. By parsing, we mean inferring the parse tree (a configuration of the proposed hierarchical model) for each face instance. In modeling, a three-layer hierarchical model is built consisting of six nodes. For each node, an active basis model is trained, and their spatial relations such as relative locations and scales are modeled using Gaussian distributions. In computing, we run the learned active basis models on testing images to obtain bottom-up hypotheses, followed by explicitly testing the compatible relations among those hypotheses to do verification and construct the parse tree in a top-down manner. In experiments, we test our approach on CMU+MIT face test set with improved performance obtained.

Geometric distortions caused by different sources usually are accumulated and not present singly in a remotely sensed image. In addition, the effects of geometric distortions are found unequally in the entire image. Hence, aerial images should be rectified before proceed with subsequent images analysis. Control points and geometric transformation are the essential components in non-parametric approach. Barrel and perspective distortions are usually found in aerial images. This paper studies the deformation rate contributed by control points at different regions in an image before rectification according to different distribution patterns. Besides, this paper also discusses the appropriate geometric transformation through the concern of the expected distortions. Experiments are conducted using grid images and aerial images to investigate the effect of distributions of control points and the efficiency of global and local geometric transformations for aerial images rectification. It demonstrated that control points at different image regions have different deformation rates, control points distributed at image centre are less distorted and local transformation performs better in rectifying images with complex distortions.

Twitter is one of the most popular social media outlets available and has been expanding over the years in both scope and reach. The growing number of users, and the accessibility to their micro-posts and metadata make Twitter a popular subject for research in various research communities. In this paper, we propose and examine a content analysis method that utilizes the hierarchy of effects model, which has a very long history of use by both practitioners and academics in the field of advertising and marketing. We have judges manually annotate tweets in accordance with one of the five stages of attitudes: Attention, Interest, Desire, Action, or Satisfaction. Next, we examine the tagged corpus to identify general traits and to explore the possibilities for the newly gained information from the tweets. The results suggest that consumer attitude information can be utilized to improve the prediction quality of box-office revenues and possibly better represent movie audience sentiment. These findings can complement other content analysis methods that utilizes Twitter data by providing an additional dimension for the researchers to consider.