Chinese Spelling Correction （CSC） is an important research task in Natural Language Processing （NLP）. The existing CSC methods based on Large Language Models （LLMs） may generate semantic discrepancies between the corrected results and the original content. Therefore， a CSC method based on LLM with multiple inputs was proposed. The method consists of two stages： multi-input candidate set construction and LLM correction. In the first stage， a multi-input candidate set was constructed using error correction results of several small models. In the second stage， LoRA （Low-Rank Adaptation） was employed to fine-tune the LLM， which means that with the aid of reasoning capabilities of the LLM， sentences without spelling errors were deduced from the multi-input candidate set and used as the final error correction results. Experimental results on the public datasets SIGHAN13， SIGHAN14， SIGHAN15 and revised SIGHAN15 show that the proposed method has the correction F1 value improved by 9.6， 24.9， 27.9， and 34.2 percentage points， respectively， compared to the method Prompt-GEN-1， which generates error correction results directly using an LLM. Compared with the sub-optimal error correction small model， the proposed method has the correction F1 value improved by 1.0， 1.1， 0.4， and 2.4 percentage points， respectively， verifying the proposed method’s ability to enhance the effect of CSC tasks.

To address the issue of insufficient extraction of semantic feature information with different scales and the lack of focus on crucial information when obtaining sentence semantic information by Convolutional Neural Network （CNN）-based relation extraction， a model for relation extraction based on a multi-scale hybrid attention CNN was proposed. Firstly， relation extraction was modeled as label prediction with two-dimensional representation. Secondly， by extracting and fusing multi-scale feature information， finer-grained multi-scale spatial information was obtained. Thirdly， through the combination of attention and convolution， the feature maps were refined adaptively to make the model concentrate on important contextual information. Finally， two predictors were used jointly to predict the relation labels between entity pairs. Experimental results demonstrate that the multi-scale hybrid convolutional attention model can capture multi-scale semantic feature information，And the key information in channels and spatial locations was captured by the channel attention and spatial attention by assigning appropriate weights， thereby improving the performance of relation extraction. The proposed model achieves F1 scores of 90.32% on SemEval （SemEval-2010 task 8） dataset， 70.74% on TACRED （TAC Relation Extraction Dataset）， 85.71% on Re-TACRED （Revised-TACRED）， and 89.66% on SciERC （Entities， Relations， and Coreference for Scientific knowledge graph construction）.

Existing methods for answer acquisition based on pre-trained language models may suffer from inaccuracies in predicting boundaries， a boundary-aware approach for span-based extraction Machine Reading Comprehension （MRC） is proposed to mitigate this issue. Firstly， special characters were introduced to mark the question boundary during the question input stage， enhancing the semantic information of the question to improve boundary perception. Secondly， during the answer prediction stage， an answer boundary regressor was constructed to facilitate semantic interaction between the perceived question boundary and the output of the predicted answer boundary. Lastly， the biased predicted answer boundary was further adjusted based on the post-interaction semantic information to calibrate the predicted answers. Experimental results demonstrate that when compared to the SpanBERT （Span-based Bidirectional Encoder Representation from Transformers）， the proposed method improves the F1 value by 0.2 percentage points and the Exact Match （EM） value by 0.9 percentage points on the public dataset SQuAD （Stanford Question Answering Dataset）1.1， it achieved improvements of 0.7 percentage points in both F1 score and EM value on the HotpotQA （Hotpot Question Answering） dataset， and it improved the F1 score by 2.8 percentage points and the EM value by 3.3 percentage points on the NewsQA （News Question Answering） dataset. The effectiveness of this method is rooted in its capacity to enhance the model’s perception of question boundary information and to accomplish the calibration of predicted answer boundary. Consequently， it results in an enhancement of system accuracy in applications such as intelligent question answering and intelligent customer service when dealing with text data comprehension and analysis.

Aiming at the problem that the existing deep clustering methods can not efficiently divide event types without considering event information and its structural characteristics， a Deep Event Clustering method based on Event Representation and Contrastive Learning （DEC_ERCL） was proposed. Firstly， information recognition was utilized to identify structured event information from unstructured text， thus the impact of redundant information on event semantics was avoided. Secondly， the structural information of the event was integrated into the autoencoder to learn the low-dimensional dense event representation， which was used as the basis for downstream clustering. Finally， in order to effectively model the subtle differences between events， a contrast loss with multiple positive examples was added to the feature learning process. Experimental results on the datasets DuEE， FewFC， Military and ACE2005 show that the proposed method performs better than other deep clustering methods in accuracy and Normalized Mutual Information （NMI） evaluation indexes. Compared with the suboptimal method， the accuracy of DEC_ERCL is increased by 17.85%，9.26%，7.36% and 33.54%， respectively， indicating that DEC_ERCL has better event clustering effect.

To tackle the difficulty in semantic mining of entity relations and biased relation prediction in Relation Extraction （RE） tasks， a RE method based on Mask prompt and Gated Memory Network Calibration （MGMNC） was proposed. First， the latent semantics between entities within the Pre-trained Language Model （PLM） semantic space was learned through the utilization of masks in prompts. By constructing a mask attention weight matrix， the discrete masked semantic spaces were interconnected. Then， the gated calibration networks were used to integrate the masked representations containing entity and relation semantics into the global semantics of the sentence. Besides， these calibrated representations were served as prompts to adjust the relation information， and the final representation of the calibrated sentence was mapped to the corresponding relation class. Finally， the potential of PLM was fully exploited by the proposed approach through harnessing masks in prompts and combining them with the advantages of traditional fine-tuning methods. The experimental results highlight the effectiveness of the proposed method. On the SemEval （SemEval-2010 Task 8） dataset， the F1 score reached impressive 91.4%， outperforming the RELA （Relation Extraction with Label Augmentation） generative method by 1.0 percentage point. Additionally， the F1 scores on the SciERC （Entities， Relations， and Coreference for Scientific knowledge graph construction） and CLTC （Chinese Literature Text Corpus） datasets were remarkable， achieving 91.0% and 82.8% respectively. The effectiveness of the proposed method was evident as it consistently outperformed the comparative methods on all three datasets mentioned above. Furthermore， the proposed method achieved superior extraction performance compared to generative methods.

With the application of artificial intelligence technology in the judicial field， charge prediction based on case description has become an important research content. It aims at predicting the charges according to the case description. The terms of case contents are professional， and the description is concise and rigorous. However， the existing methods often rely on text features， but ignore the difference of relevant elements and lack effective utilization of elements of action words in diverse cases. To solve the above problems， a multi-task learning model of charge prediction based on action words was proposed. Firstly， the spans of action words were generated by boundary identifier， and then the core contents of the case were extracted. Secondly， the subordinate charge was predicted by constructing the structure features of action words. Finally， identification of action words and charge prediction were uniformly modeled， which enhanced the generalization of the model by sharing parameters. A multi-task dataset with action word identification and charge prediction was constructed for model verification. The experimental results show that the proposed model achieves the F value of 83.27% for action word identification task， and the F value of 84.29% for charge prediction task； compared with BERT-CNN， the F value respectively increases by 0.57% and 2.61%， which verifies the advantage of the proposed model in identification of action words and charge prediction.

It is importantly used in the fields such as creation of large-scale professional talent pools to extract scholar fine-grained information such as scholar’s research directions， education experience from scholar homepages. To address the problem that the existing scholar fine-grained information extraction methods cannot use contextual semantic associations effectively， a scholar fine-grained information extraction method incorporating local semantic features was proposed to extract fine-grained information from scholar homepages by using semantic associations in the local text. Firstly， general semantic representation was learned by the full-word mask Chinese pre-trained model RoBERTa-wwm-ext. Subsequently， the representation vector of the target sentence， as well as its locally adjacent text representation vector from the general semantic embeddings， were jointly fed into a CNN （Convolutional Neural Network） to accomplish local semantic fusion， thereby obtaining a higher-dimensional representation vector for the target sentence. Finally， the representation vector of the target sentence was mapped from the high-dimensional space to the low-dimensional labeling space to extract the fine-grained information from the scholar homepage. Experimental results show that the micro-average F1 score of the scholar fine-grained information extraction method fusing local semantic features reaches 93.43%， which is higher than that of RoBERTa-wwm-ext-TextCNN method without fusing local semantic by 8.60 percentage points， which verifies the effectiveness of the proposed method on the scholar fine-grained information extraction task.

The development of hot news events is very rich， and each stage of the development has its own unique narrative. With the development of events， a trend of hierarchical storyline evolution is presented. Aiming at the problem of poor interpretability and insufficient hierarchy of storyline in the existing storyline generation methods， a Hierarchical Storyline Generation Method （HSGM） for hot news events was proposed. First， an improved hotword algorithm was used to select the main seed events to construct the trunk. Second， the hotwords of branch events were selected to enhance the branch interpretability. Third， in the branch， a storyline coherence selection strategy fusing hotword relevance and dynamic time penalty was used to enhance the connection of parent-child events， so as to build hierarchical hotwords， and then a multi-level storyline was built. In addition， considering the incubation period of hot news events， a hatchery was added during the storyline construction process to solve the problem of neglecting the initial events due to insufficient hotness. Experimental results on two real self-constructed datasets show that in the event tracking process， compared with the methods based on singlePass and k-means respectively， HSGM has the F score increased by 4.51% and 6.41%， 20.71% and 13.01% respectively； in the storyline construction process， HSGM performs well in accuracy， comprehensibility and integrity on two self-constructed datasets compared with Story Forest and Story Graph.

The rapid development of Internet leads to the explosive growth of news data. How to capture the topic evolution process of current popular events from massive news data has become a hot research topic in the field of document analysis. However， the commonly used traditional dynamic clustering models are inflexible and inefficient when dealing with large-scale datasets， while the existing deep document clustering models lack a general method to capture the topic evolution process of time series data. To address these problems， a Deep Dynamic Document Clustering （DDDC） model was designed. In this model， based on the existing deep variational inference algorithms， the topic distributions incorporating the content of previous time slices on different time slices were captured， and the evolution process of event topics was captured from these distributions through clustering. Experimental results on real news datasets show that compared with Dynamic Topic Model （DTM）， Variational Deep Embedding （VaDE） and other algorithms， DDDC model has the clustering accuracy and Normalized Mutual Information （NMI） improved by at least 4 percentage points averagely and at least 3 percentage points respectively in each time slice on different datasets， verifying the effectiveness of DDDC model.

In recent years， due to the advantages of the structural information of Graph Neural Network （GNN） in machine learning， people have begun to combine GNN into deep text clustering. The current deep text clustering algorithm combined with GNN ignores the important role of the decoder on semantic complementation in the fusion of text semantic information， resulting in the lack of semantic information in the data generation part. In response to the above problem， a Structured Deep text Clustering Model based on multi-layer Semantic fusion （SDCMS） was proposed. In this model， a GNN was utilized to integrate structural information into the decoder， the representation of text data was enhanced through layer-by-layer semantic complement， and better network parameters were obtained through triple self-supervision mechanism.Results of experiments carried out on 5 real datasets Citeseer， Acm， Reutuers， Dblp and Abstract show that compared with the current optimal Attention-driven Graph Clustering Network （AGCN） model， SDCMS in accuracy， Normalized Mutual Information （NMI ） and Average Rand Index （ARI） has increased by at most 5.853%， 9.922% and 8.142%.

Aiming at the problem of poor recognition of sentencing circumstances in adjudication documents caused by the lack of labeled data， low quality of labeling and existence of strong logicality in judicial field， a sentencing circumstance recognition model based on abductive learning named ABL-CON （ABductive Learning in CONfidence） was proposed. Firstly， combining with neural network and domain logic inference， through the semi-supervised method， a confidence learning method was used to characterize the confidence of circumstance recognition. Then， the illogical error circumstances generated by neural network of the unlabeled data were corrected， and the recognition model was retrained to improve the recognition accuracy. Experimental results on the self-constructed judicial dataset show that the ABL-CON model using 50% labeled data and 50% unlabeled data achieves 90.35% and 90.58% in Macro_F1 and Micro_F1， respectively， which is better than BERT （Bidirectional Encoder Representations from Transformers） and SS-ABL （Semi-Supervised ABductive Learning） under the same conditions， and also surpasses the BERT model using 100% labeled data. The ABL-CON model can effectively improve the logical rationality of labels as well as the recognition ability of labels by correcting illogical labels through logical abductive correctness.

Relation extraction aims to extract the semantic relationships between entities from the text. As the upper-level task of relation extraction， entity recognition will generate errors and spread them to relation extraction， resulting in cascading errors. Compared with entities， entity boundaries have small granularity and ambiguity， making them easier to recognize. Therefore， a relationship extraction method based on entity boundary combination was proposed to realize relation extraction by skipping the entity and combining the entity boundaries in pairs. Since the boundary performance is higher than the entity performance， the problem of error propagation was alleviated； in addition， the performance was further improved by adding the type features and location features of entities through the feature combination method， which reduced the impact caused by error propagation. Experimental results on ACE 2005 English dataset show that the proposed method outperforms the table-sequence encoders method by 8.61 percentage points on Macro average F1-score.

Aiming at the problems that the existing Deep Convolutional Neural Network (DCNN) have insufficient defect image feature extraction ability, few recognition types and long reasoning time in hydraulic tunnel defect recognition tasks, an autonomous defect recognition method based on dynamic feature distillation was proposed. Firstly, the deep curve estimation network was used to optimize the image to improve the image quality in low illumination environment. Secondly, the dynamic convolution module with attention mechanism was constructed to replace the traditional static convolution, and the obtained dynamic features were used to train the teacher network to obtain better model feature extraction ability. Finally, a dynamic feature distillation loss was constructed by fusing the discriminator structure in the knowledge distillation framework, and the dynamic feature knowledge was transferred from the teacher network to the student network through the discriminator, so as to achieve the high-precision recognition of six types of defects while significantly reducing the model reasoning time. In the experiments, the proposed method was compared with the original residual network on a hydraulic tunnel defect dataset of a hydropower station in Sichuan Province. The results show that this method has the recognition accuracy reached 96.15%, and the model parameter amount and reasoning time reduced to 1/2 and 1/6 of the original ones respectively. It can be seen from the experimental results that fusing the dynamic feature distillation information of the defect image into the recognition network can improve the efficiency of hydraulic tunnel defect recognition.

Aiming at the safety and smoothness problems of path planning for Unmanned Surface Vehicle (USV) in complex marine environment, a multi-directional A ^* path planning algorithm was proposed for obtaining global optimal path. Firstly, combining the electronic chart, the rasterized environment information was established, and a safe area model of USV was established according to the safety nevigation distance constraint. And the A ^* heuristic function with safety distance constraint was designed based on the traditional A ^* algorithm to ensure the safety of the generated path nodes. Secondly, a multi-directional search mode was proposed by improving the eight-directional search mode of the traditional A ^* algorithm to adjust the redundant points and inflection points in the generated path. Finally, the path smoothing algorithm was used to smooth the inflection points to obtain the continuous smooth path that meets the actual navigation requirements. In the simulation experiment, the path distance planned by the improved A ^* algorithm is 7 043 m, which is 9.7%, 26.6% and 7.9% lower than those of Dijkstra algorithm, traditional A ^* four-directional search algorithm and traditional A ^* eight-directional search algorithm. The simulation results show that the improved multi-directional A ^* search algorithm can effectively reduce the path distance, and is more suitable for the path planning of USV.

Concerning the problems that the traditional Convolutional Neural Network (CNN) ignores the contextual semantics of words when performing sentiment analysis tasks and CNN loses a lot of feature information during max pooling operation at the pooling layer, which limit the text classification performance of model, a parallel hybrid neural network model, namely CA-BGA (Convolutional Neural Network Attention and Bidirectional Gated Recurrent Unit Attention), was proposed. Firstly, a feature fusion method was adopted to integrate Bidirectional Gated Recurrent Unit (BiGRU) into the output of CNN, thus semantic learning was enhanced by integrating the global semantic features of sentences. Then, the attention mechanism was introduced between the convolutional layer and the pooling layer of CNN and at the output of BiGRU to reduce noise interference while retaining more feature information. Finally, a parallel hybrid neural network model was constructed based on the above two improvement strategies. Experimental results show that the proposed hybrid neural network model has the characteristic of fast convergence, and effectively improves the F1 value of text classification. The proposed model has excellent performance in Chinese short text sentiment analysis tasks.

In order to solve the problem that the approximator of the current approximation policy iteration reinforcement learning cannot be constructed completely automatically, a reinforcement learning algorithm of Nonparametric Approximation Policy Iteration based on Dyna Framework (NPAPI-Dyna) was proposed. Sampling cache and sampling change rate were introduced to design a two stage random sampling process to collect samples. By profile tolerance and K-means clustering, core state basis function was generated through trial-and-error process. Q-value function approximator was generated by using the complete coverage of sample as the target. Greedy strategy was applied to design action selector. Access frequency of the state basis function was used to describe environmental topology features and construct environment estimation model. Learning and planning processes were combined organically by identification of Dyna framework to accelerate the speed of learning.In the simulation experiments of single inverted pendulum balance control, when the reinforcement learning error rate is 0.01, the learning success rate of algorithm reaches 100%, the minimum number of successful attempts is only 2, the average number of attempts is only 7.73, and the mean absolute deviation of angle is 3.0538°, and the average oscillation range of angle is 2.759°. When reinforcement learning error rate is 0.1, 100 independent simulation operations are performed, to learn the control strategy, Online-LSPI and BLSPI (Batch Least-Squares Policy Iteration) have to try more than 150 times on average, however NPAPI-Dyna can succeed in 50 times of attempts. The experimental results show that NPAPI-Dyna can be completely automatically constructed and adjusted to enhance the learning structure, with high learning accuracy and rapid convergence ability.

Aiming at the problem of the poor real-time performance of Inertial Navigation System (INS)/Global Positioning System (GPS) integrated navigation system caused by GPS information delay, a processing method which takes advantage of dealing with various asynchronous measurements at an information fusion time in factor graph algorithm was proposed. Before the system received GPS information, the factor nodes of the INS information were added to the factor graph model, and the integrated navigation results were obtained by incremental inference to ensure the real-time performance of the system. After the system received the GPS information, the factor nodes about the GPS information were added to the factor graph model to correct the INS error, thereby ensuring high-precision operation of the system for a long time. The simulation results show that, the navigation state that has just been updated by GPS information can correct the INS error effectively, when the correction effect of real-time navigation state on INS error becomes worse, as the time of GPS information delay becomes longer. The factor graph algorithm avoids the adverse effects of GPS information delay on the real-time performance of INS/GPS integrated navigation system, and ensures the accuracy of the system.

Focused on the problem of low recognition in recognition systems caused by the inaccuracy of input features, an efficient Deep Sparse Auto-Encoder (DSAE) method using Extreme Learning Machine (ELM) for facial features was proposed. Firstly, truncated nuclear norm was used to construct loss function, and sparse features of face images were extracted by minimizing loss function. Secondly, self-encoding of facial features was used by Extreme Learning Machine Auto-Encoder (ELM-AE) model to achieve data dimension reduction and noise filtering. Thirdly, the optimal depth structure was obtained by minimizing the empirical risk. The experimental results on ORL, IMM, Yale and UMIST datasets show that the DSAE method not only has higher recognition rate than ELM, Random Forest (RF), etc. on high-dimensional face images, but also has good generalization performance.

For such difficulties as low mining efficiency and large amount of data that the data mining of residential electricity data has to be faced with, the analysis based on improved K-means algorithm and cloud computing on massive data of power utilization was researched. As the initial cluster center and the value K are difficult to determine in traditional K-means algorithm, an improved K-means algorithm based on density was proposed. Firstly, the product of sample density, the reciprocal of the average distance between the samples in the cluster, and the distance between the clusters were defined as weight product, the initial center was determined successively according to the maximum weight product method and the accuracy of the clustering was improved. Secondly, the parallelization of improved K-means algorithm was realized based on MapReduce model and the efficiency of clustering was improved. Finally, the mining experiment of massive power utilization data was carried out on the basis of 400 households' electricity data. Taking a family as a unit, such features as electricity consumption rate during peak hour, load rate, valley load coefficient and the percentage of power utilization during normal hour were calculated, and the feature vector of data dimension was established to complete the clustering of similar user types, at the same time, the behavioral characteristics of each type of users were analyzed. The experimental results on Hadoop cluster show that the improved K-means algorithm operates stably and efficiently and it can achieve better clustering effect.

The problems of face detection in bus environment include ambient illumination changing, image distortion, human body occlusion, abnormal postures and etc. For alleviating these mentioned limitations, a face detection based on cost-sensitive Deep Quadratic Tree (DQT) was proposed. First of all, Normalized Pixel Difference (NPD) feature was utilized to construct and train a single DQT. According to the classification result of the current decision tree, the cost-sensitive Gentle Adaboost method was used to update the sample weight, and a number of deep decision trees were trained. Finally, the classifier was produced by Soft-Cascade method with multiple upgraded deep quadratic trees. The experimental results on Face Detection Data set and Benchmark (FDDB) and bus video show that compared with the existing depth decision tree algorithm, the proposed algorithm has improved the detection rate and detection speed.

A new Kinect depth map pretreatment algorithm was presented for the lower accuracy problem compared with the original depth information in the field of three-Dimensional (3D) scene measurement for robot's perception. Firstly, a measuring and sampling model of the depth map was developed to realize the Monte Carlo uncertainty evaluation model. Secondly, the depth value intervals were calculated to judge and filter the noise pixels. Finally, noise points were repaired with mean-value of the estimation intervals. The experimental results show that the algorithm can effectively suppress and repair the noise pixels while keeping the depth gradient and values of non-noise pixels. The Mean Square Error (MSE) of depth map after preprocessing is reduced by 15.25% to 28.79%, and the object profiles remain unchanged compared with the JBF (Joint Bilateral Filtering) based on color and depth map. Therefore, it achieves the purpose of improving the depth information accuracy in 3D scenes.

Many methods for liver tumor Computed Tomography (CT) segmentation have the difficulty to achieve accurate tumor due to inhomogeneous gray and fuzzy edges. To obtain precise segmentation result, a method using multi-scale morphology was proposed to eliminate local minima. Firstly, the morphological area operation was used to remove image's small burrs and irregular edges so as to avoid boundaries migration. Secondly, local minima in gradient image were distinguished by the combined knowledge of statistic characteristics and morphological properties including depth and scale. After partition, the function relationship was established between multi-scale structure elements and local minima. In order to filter noise via large-size structure elements and preserving major object via small-size structure elements, a morphological method called close operation was then employed to adaptively modify the image.Finally, standard watershed transform was utilized to implement segmentation of liver tumor. The experimental results show that this method can reduce over-segmentation effectively and liver tumor can be segmented accurately while boundaries of objects are located precisely.

Focusing on the key exchange problem of how to get the higher security for neural cryptography in the short time of the synchronization, a new hybrid algorithm combining the features of "Do not Trust My Partner" (DTMP) and the fast learning rule was proposed. The algorithm could send erroneous output bits in the public channel to disrupt the attacker's eavesdropping of the exchanged bits and reduce the success rate of passive attack. Meanwhile, the proposed algorithm estimated the synchronization by estimating the probability of unequal outputs, then adjusted the change of weights according to the level of synchronization to speed up the process of synchronization. The simulation results show that the proposed algorithm outperforms the original DTMP in the time needed for the partners to synchronize. Moreover, the proposed algorithm is securer than the original DTMP when the partners do not send erroneous output bits at the same time. And the proposed algorithm outperforms the feedback algorithm in both the synchronization time and security obviously. The experimental results show that the proposed algorithm can obtain the key with a high level of security and a less synchronization time.

Since the clustering by data competition algorithm has poor performance on complex datasets, a density-sensitive clustering by data competition algorithm was proposed. Firstly, the local distance was defined based on density-sensitive distance measure to describe the local consistency of data distribution. Secondly, the global distance was calculated based on local distance to describe the global consistency of data distribution and dig the information of data space distribution, which can make up for the defect of Euclidean distance on describing the global consistency of data distribution. Finally, the global distance was used in clustering by data competition algorithm. Using synthetic and real life datasets, the comparison experiments were conducted on the proposed algorithm and the original clustering by data competition based on Euclidean distance. The simulation results show that the proposed algorithm can obtain better performance in clustering accuracy rate and overcome the defect that clustering by data competition algorithm is difficult to handle complex datasets.

Regional stereo matching algorithm is sensitive to illumination change. Disparity map has the problems that target and the weak texture region mismatch, boundary is not smooth, and so on. In order to solve these problems, an improved quick stereo matching algorithm by using visual saliency characteristics was proposed. Saliency detection was used to locate the main target area in the image. Then feature matching was completed by combining image Sobel edge characteristics and phase features to get the rough disparity map. Finally, by detecting visual saliency in the disparity map, abrupt noise in weak texture image area was eliminated. Compared to the traditional algorithms such as Sum of Absolute Differences (SAD), Sum of Squared Differences (SSD), Normalized Cross Correlation (NCC), the proposed algorithm is not sensitive to light changing, and it can get better disparity map and higher matching rate, which is conducive to real-time applications.

In order to improve the parameter optimization results of cognitive wireless network, an immune optimization based parameter adjustment algorithm was proposed. Engine parameter adjustment of cognitive wireless network is a multi-objective optimization problem. Intelligent optimization method is suitable for solving it. Immune clonal optimization is an effective intelligent optimization algorithm. The mutation probability affects the searching capabilities in immune optimization. Cloud droplets have randomness and stable tendency in normal cloud model, so an adaptive mutation probability adjustment method based on cloud model was proposed, and it was used in parameter optimization of cognitive radio networks. The simulation experiments were done to test the algorithm under multi-carrier system. The results show that, compared with relative algorithms, the proposed algorithm has better convergence, and the parameter adjustment results are consistent with the preferences for the objectives function. It can get optimal parameter results of cognitive engine.

Concerning the issue of how to achieve effective estimation of gravitational acceleration for attitude estimation of Micro Air Vehicle (MAV) under all dynamic conditions, an improved explicit complementary filter was proposed in combination with stepped-gain schedule. In order to validate the nonlinear complementary filter in the case when a MAV circled for an extended period, a centripetal acceleration mode was built using gyroscopes and indicated airspeed data, which resulted in precise estimation based on the estimation of gravitational acceleration and avoided reconstructing an estimation of the attitude. In the phase of Proportional-Integral (PI) compensation, the proportional gain and integral gain can achieve better adaptability by assigning different cut-off frequency value to the estimation of pitch and roll angles respectively. The experimental results show that the attitude angle estimation can be maintained under the range of ±2°. Compared with the typical filter algorithm, a better performance was achieved with respect to the efficiency and estimation error, so the algorithm proposed in this paper can be applied to accurate attitude estimation for MAV with low-cost inertial measurement units.

In order to meet the needs of precise positioning of the mobile robot, a fingerprint positioning method of particle filter based on different resampling algorithms was presented. Firstly, during the positioning phase, the motion model built on robot kinematics served as the proposal density of particle filter, and the observation information and environment fingerprint were infused into the filtering process to enhance the particles' refining capacity and reduce the required number of particles. Secondly, an Exquisite Resampling (ER) algorithm was introduced to improve the refining ability of the particles, thus the effect of particle impoverishment could be decreased and the localization accuracy could be improved. At last, the influence of the positioning accuracy caused by different re-sampling algorithms was analyzed, and a further investigation on the accuracy and reliability of localization algorithm from different experimental perspectives was given. The experimental results show that this algorithm has the advantages in localization accuracy and robustness.

As microblog application grows rapidly, how to extract users' interested popular topic from massive microblog information automatically becomes a challenging research area. This paper studied and proposed a topic extraction algorithm of Chinese microblog based on extended topic model. In order to deal with data sparse problem of microblog, the content related microblog text would be firstly clustered to generate synthetic document. Based on the assumption that posting relationship among microblogs implied topical correlation, the traditional LDA (Latent Dirichlet Allocation) topic model was extended to model the posting relationship among microblogs. At last, Mutual Information (MI) measurement was utilized to calculate topic vocabulary after extracting topics by proposing extended LDA topic model for topic recommendation. Furthermore, a prototype system for domain-specific topical mining system, named BTopicMiner, was implemented so as to verify the effectiveness of the proposed algorithm. The experimental result shows that the proposed algorithm can extract topics from microblogs more accurately. Meanwhile, the semantic similarity between automatically calculated topic vocabulary and manually selected topic vocabulary exceeds 75% while automatically calculating topic vocabulary based on MI.

Social annotation is a form of folksonomy, which allows Web users to categorize Web resource with text tags freely. It usually implicates fundamental and valuable semantic information of Web resources. Consequently, social annotation is helpful to improve the quality of information retrieval when applied to information retrieval system. This paper investigated and proposed an improved text classification algorithm based on social annotation. Because social annotation is a kind of folksonomy and social tags are usually generated arbitrarily without any control or expertise knowledge, there has been significant variance in the quality of social tags. Under this consideration, the paper firstly proposed a quantitative approach to measure the quality of social tags by utilizing the semantic similarity between Web pages and social tags. After that, the social tags with relatively low quality were filtered out based on the quality measurement and the remained social tags with high quality were applied to extend traditional vector space model. In the extended vector space model, a Web page was represented by a vector in which the components were the words in the Web page and tags tagged to the Web page. At last, the support vector machine algorithm was employed to perform the classification task. The experimental results show that the classification result can be improved after filtering out the social tags with low quality and embedding those high quality social tags into the traditional vector space model. Compared with other classification approaches, the classification result of F1 measurement has increased by 6.2% on average when using the proposed algorithm.