At present, the accuracy of image retrieval is a difficult problem to study, the main reason is the method of feature extraction. In order to improve the precision of image retrieval, a new image retrieval method based on multi-feature called CAUC (Comprehensive Analysis based on the Underlying Characteristics) was presented. First, based on YUV color space, the mean value and the standard deviation were used to extract the global feature from an image that depicted the global characteristics of the image, and the image bitmap was introduced to describe the local characteristics of the image. Secondly, the compactness and Krawtchouk moment were extracted to describe the shape features. Then, the texture features were described by the improved four-pixel co-occurrence matrix. Finally, the similarity between images was computed based on multi-feature fusion, and the images with high similarity were returned.On Corel-1000 image set, the comparative experiments with method which only considered four-pixel co-occurrence matrix showed that the retrieval time of CAUC was greatly reduced without significantly reducing the precision and recall. In addition, compared with the other two kinds of retrieval methods based on multi-feature fusion, CAUC improved the precision and recall with high retrieval speed. The experimental results demonstrate that CAUC method is effective to extract the image feature, and improve retrieval efficiency.
To solve the problem that present traversal methods of hierarchical tree which lead to low efficiency, a new collision detection algorithm based on classified traversal was proposed. Firstly, these objects were classified according to the difference between the balance factors of two tree' nodes. The simultaneous depth-first traversal method was applied to the objects which have similar structure, and the commutative depth-first traversal method was applied to the other objects, which reduced the number of intersect tests. Then, the process of traversal was optimized by using the temporal spatial coherence and priority strategy. Finally, the experimental results show that, compared with the collision detection algorithm based on unified traversal, the proposed algorithm shortens the time of the intersection test. The bigger the number of objects, the more significant the advantage of quickness, it can reduce about 1/5 of the required time.
To solve the slow rendering problem of ray tracing algorithm in irregular scene, an improved grid subdivision ray tracing algorithm was proposed based on the in-depth study and comparison of the recent acceleration algorithms of ray tracing. First, the rectangular bounding box of the scene was set to remove the influence of external light, and the intersect operations was simplified; second, spatial grids were created with a new way to limit the spatial unit number and complexity of storage space within a certain range; finally, the traditional spatial grid algorithm was greatly improved by subdividing grids to eliminate the bad influence on acceleration effects due to ignoring some blank space. The experimental results show that this method can effectively improve the light speed in blank space, it not only increases the time efficiency but also reduces the space lost.
Concerning the problem that how to initialize the weights of deep neural networks, which resulted in poor solutions with low generalization for spam filtering, a classification method of Deep Belief Net (DBN) was proposed based on the fact that the existing spam classifications are shallow learning methods. The DBN was pre-trained with the greedy layer-wise unsupervised algorithm, which achieved the initialization of the network. The experiments were conducted on three datesets named LinsSpam, SpamAssassin and Enron1. It is shown that compared with Support Vector Machines (SVM) which is the state-of-the-art method for spam filtering in terms of classification performance, the spam filtering using DBN is feasible, and can get better accuracy and recall.
In order to improve the speed and accuracy of image retrieval, the drawbacks of image retrieval based on a variety of clustering algorithms were analyzed, then a new partition clustering method for image retrieval was presented in this paper. First, based on the asymmetrical quantization of the color in HSV model, color feature of image was extracted by color coherence vectors. Then, global shape feature of image was extracted based on improved Hu invariant moment. Finally,images were clustered based on contribution according to color and shape features, and image feature index library was established. The methods described above were used for image retrieval based Corel image library. The experimental results show that compared with image retrieval algorithms based on improved K-means algorithms, precision ratio and recall ratio of the proposed algorithm are improved greatly.