To solve the problem of k nearest neighbor query in obstacle space, a k nearest neighbor Query method based on improved Parallel Ant colony algorithm (PAQ) was proposed. Firstly, ant colonies with different kinds of pheromones were utilized to search k nearest neighbors in parallel. Secondly, a time factor was added as a condition of judging path length to directly show the searching time of ants. Thirdly, the concentration of initial pheromone was redefined to avoid the blind searching of ants. Finally, visible points were introduced to divide the obstacle path into multiple Euclidean paths, meawhile the heuristic function was improved and the visible points were selected by ants to conduct probability transfer making ants search in more proper direction and prevent the algorithm from falling into local optimum early. Compared to WithGrids method, with number of data points less than 300, the running time for line segment obstacle is averagely reduced by about 91.5%, and the running time for polygonal obstacle is averagely reduced by about 78.5%. The experimental results show that the running time of the proposed method has obvious advantage on small-scale data, and the method can process polygonal obstacles.

Conventional watermarking algorithms usually pay more attention to the visual quality of embedded carrier while ignore the security of watermarking. Although some methods provided watermarking encryption procedures, they usually embed watermarks in fixed positions which are prone to be attacked. The sensitivity of watermarking algorithm based on parameterized wavelet transform is difficult to be applied in practice. To address these problems, a public sensitive watermarking algorithm with weighted multilevel wavelet coefficient mean and quantization was proposed. In the proposed algorithm, firstly the Message Digest Algorithm 5 (MD5) value of cover image, user keys and initial parameters were bound with Logistic map which were used to encrypt watermarks and select wavelet coefficients in different decomposition levels; secondly weights of wavelet coefficients in different levels were estimated by absolute variation means of wavelet coefficients before and after Joint Photographic Experts Group (JPEG) compression, and then weighted multilevel wavelet coefficient mean was adjusted to embed watermark; finally an isolated black point filtering strategy was adopted to enhance the quality of fetched watermark. The experiments show the proposed method has better sensitivities of plaintext image and user keys and still is robust for common image attacks such as image clipping, white noise, JPEG compression, covering and graffiti. The Peak Signal-to-Noise Ratio (PSNR) of image after embedding watermarks can reach 45 dB. The embedded watermark is difficult to be tampered or extracted even if all watermarks embedding procedures are published.

The analysis about Argo buoy profile is not comprehensive for taking the single Argo buoy as a processing object, and the calculation time of uniprocessing method is long. In order to solve the problems, a new algorithm using latitude and longitude cell as analysis object and combining MapReduce with principal curve analysis was proposed. In Map processing, the effective information of Argo buoy was extracted from the big data files and the extracted Argo profiles were classified according to the latitude and longitude. In Reduce processing, the principal Argo profile of each region was generated. Firstly, the information was normalized, and then the principal Argo profile of regional profile characteristics which consisted of a small amount of profile points and lines was obtained through the Kegl's principal curve theory, the information fusion of massive Argo buoy was realized. The proposed algorithm was verified through the global Argo buoy sample data, the new algorithm achieved the mean of residual errors was within 0.1 in the condition of 0.03-0.10 squared distance, the data storage space was saved by 99.4%, and the computation speed was increased by 36.4%, compared with the traditional method only based on uniprocessing. The experimental results show that the proposed algorithm can generate principal profiles accurately, meanwhile reduce the data storage space and effectively improve the computation speed.

Concerning that the lesion areas of leukoaraiosis in Magnetic Resonance (MR) image present hyper intense signal on T ₂ flair sequence, a level set segmentation method based on C-V model was proposed. First, the C-V model was improved to avoid the re-initialization; second, the Otsu threshold method was used for image's pre-segmentation, and then the image's pre-segmentation result was directly used as the initial contour for the improved C-V model; finally, the segmentation result was obtained by curve evolution. The results show that the proposed segmentation method can get better separation effects, and realize fast auto-segmentation. It has certain application value for clinical diagnosis and prognosis on leukoaraiosis.

Aiming at the lack of the function of 3D Modeling of ComGIS,the idea of 3D Modeling based on ComGIS and CAD’ Integration was put forward,and an integration method of 3D Modeling was introduced.The idea was that GIS data’ import,3D Modeling and raster picture of 3D perspective drawing were realized by Visual LISP programming of AutoCAD,the command of WIN32 API ran AutoCAD in background,and Command Scripts realized loading and running of Visual LISP,and at last,GIS displayed 3D Model.

 Automatic interpretation of synthetic aperture radar (SAR) images was one of the most important application fields in image processing. Focusing on the medium resolution SAR images and combining with the previous algorithms, a novel technique to detecting ship targets from coastal regions based on fuzzy theory was proposed. In this new method, the input image was first processed with improved fuzzy enhancement algorithm so as to alter the characteristics of the gray-level distribution; then, with the threshold method, the sea and land regions could be separated; following that, the maximum entropic algorithm was employed to further process the image and regions of interest which contained candidate ship targets could be extracted; Then ROIs were segmented and the features of ship targets were extracted. Finally the ship targets could be detected based on fuzzy reasoning technique.

Aimed at the inter-block dependency, an image classification algorithm based on a two hidden Markov model(2DHMM) extension from the one dimensional HMM was developed. The 2DHMM has transition probabilities conditioned on the states of neighboring blocks from both directions. Thus, the dependency in two dimensions can be reflected simultaneously. The HMM parameters were estimated by the EM algorithm. A two dimensional version of the Viterbi algorithm was also developed to classify optimally an image based on the trained HMM. Application of the HMM algorithm to document image shows that the algorithm performs better than CART.