scholarly journals A Novel SAR Image Target Recognition Algorithm under Big Data Analysis

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Xiang Chen ◽  
Xing Wang ◽  
You Chen ◽  
Haihan Wang
Keyword(s):  
Big Data ◽  
Data Analysis ◽  
Target Recognition ◽  
Image Interpretation ◽  
Image Data ◽  
Automatic Target Recognition ◽  
Big Data Analysis ◽  
Recognition Algorithm ◽  
Sar Image ◽  
Data Set

Synthetic aperture radar (SAR) image target recognition technology is aimed at automatically determining the presence or absence of target information from the input SAR image and improving the efficiency and accuracy of SAR image interpretation. Based on big data analysis, dirty data is removed, clean data is returned, and standardized processing of SAR image data is realized. At the same time, by establishing a statistical model of coherent speckles, the convolutional autoencoder is used to denoise the SAR image. Finally, the network model modified by softmax cross-entropy loss and Fisher loss is used for automatic target recognition. Based on the MSTAR data set, two scene graphs containing the target synthesized by the background image and the target slice are used for experiments. Several comparative experiments have verified the effectiveness of the classification and recognition model in this paper.

scholarly journals Robust Automatic Target Recognition Algorithm for Large-Scene SAR Images and Its Adaptability Analysis on Speckle

2016 ◽  
Vol 2016 ◽  
pp. 1-11 ◽  
Author(s):  
Hongqiao Wang ◽  
Yanning Cai ◽  
Guangyuan Fu ◽  
Shicheng Wang
Keyword(s):  
Target Recognition ◽  
Low Frequency ◽  
Automatic Target Recognition ◽  
Recognition Algorithm ◽  
Scale Transformation ◽  
Sar Image ◽  
Sar Images ◽  
Stationary Target ◽  
Multilevel Sampling ◽  
Process Method

Aiming at the multiple target recognition problems in large-scene SAR image with strong speckle, a robust full-process method from target detection, feature extraction to target recognition is studied in this paper. By introducing a simple 8-neighborhood orthogonal basis, a local multiscale decomposition method from the center of gravity of the target is presented. Using this method, an image can be processed with a multilevel sampling filter and the target’s multiscale features in eight directions and one low frequency filtering feature can be derived directly by the key pixels sampling. At the same time, a recognition algorithm organically integrating the local multiscale features and the multiscale wavelet kernel classifier is studied, which realizes the quick classification with robustness and high accuracy for multiclass image targets. The results of classification and adaptability analysis on speckle show that the robust algorithm is effective not only for the MSTAR (Moving and Stationary Target Automatic Recognition) target chips but also for the automatic target recognition of multiclass/multitarget in large-scene SAR image with strong speckle; meanwhile, the method has good robustness to target’s rotation and scale transformation.

Research on the Data Driven Practice Teaching Mode: Take the Didi Data Set as Example

2021 ◽  
Vol 105 ◽  
pp. 348-355
Author(s):  
Hou Xiang Liu ◽  
Sheng Han Zhou ◽  
Bang Chen ◽  
Chao Fan Wei ◽  
Wen Bing Chang ◽  
...  
Keyword(s):  
Big Data ◽  
Data Analysis ◽  
Teaching Method ◽  
Rapid Development ◽  
Big Data Analysis ◽  
Theoretical Knowledge ◽  
Data Set ◽  
Teaching Mode ◽  
Practice Teaching ◽  
Practical Teaching

The paper proposed a practice teaching mode by making analysis on Didi data set. There are more and more universities have provided the big data analysis courses with the rapid development and wide application of big data analysis technology. The theoretical knowledge of big data analysis is professional and hard to understand. That may reduce students' interest in learning and learning motivation. And the practice teaching plays an important role between theory learning and application. This paper first introduces the theoretical teaching part of the course, and the theoretical methods involved in the course. Then the practice teaching content of Didi data analysis case was briefly described. And the study selects the related evaluation index to evaluate the teaching effect through questionnaire survey and verify the effectiveness of teaching method. The results show that 78% of students think that practical teaching can greatly improve students' interest in learning, 89% of students think that practical teaching can help them learn theoretical knowledge, 89% of students have basically mastered the method of big data analysis technology introduced in the course, 90% of students think that the teaching method proposed in this paper can greatly improve students' practical ability. The teaching mode is effective, which can improve the learning effect and practical ability of students in data analysis, so as to improve the teaching effect.

scholarly journals Demand Analysis of Online Chinese Behavior Expression in Wireless Sensor Network

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Zheng Liu
Keyword(s):  
Wireless Sensor Networks ◽  
Big Data ◽  
Data Analysis ◽  
Wireless Sensor Network ◽  
Sensor Networks ◽  
Semantic Analysis ◽  
Big Data Analysis ◽  
Wireless Sensor ◽  
Data Set ◽  
Analysis Models

Due to the common progress and interdependence of wireless sensor networks and language, Chinese semantic analysis under wireless sensor networks has become more and more important. Although there are many research results on wireless networks and Chinese semantics, there are few researches on the influence and relationship between them. Wireless sensor networks have strong application relevance, and the key technologies that need to be solved are also different for different application backgrounds. In order to reveal the basic laws and development trends of online Chinese semantic behavior expression in the context of wireless sensor networks, this paper adopts big data analysis methods and semantic model analysis methods and constructs semantic analysis models through PLSA method calculations, so that the λ construction process conforms to this research topic. Research the accuracy and applicability of the semantic analysis model. Through word extraction of 1.05 million word data of 1,103 documents on Baidu Tieba, HowNet, and citeulike websites, the data set was integrated into a data set, and the PLSA model was verified with this data set. In addition, through the construction of the wireless sensor network, the semantic analysis results in the expression of Chinese behavior are obtained. The results show that the accuracy of the data set extracted from 1103 documents increases with the increase of the number of documents. Second, after using the PLSA model to perform semantic analysis on the data set, the accuracy of the data set is improved. Compared with traditional semantic analysis, the model and the big data analysis framework have obvious advantages. With the continuous development of Internet big data, the big data methods used to count Chinese semantics are also constantly updated, and their efficiency is constantly improving. These updated semantic analysis models and statistical methods are constantly eliminating the uncertainty of modern online Chinese. The basic laws and development trends of statistical Chinese semantics also provide new application scenarios for online Chinese behavior. It also laid a ladder for subsequent scholars.

Big data analysis with artificial intelligence technology based on machine learning algorithm

2020 ◽  
Vol 39 (5) ◽  
pp. 6733-6740
Author(s):  
Zeliang Zhang
Keyword(s):  
Machine Learning ◽  
Big Data ◽  
Data Analysis ◽  
Statistical Data ◽  
Big Data Analysis ◽  
Data Set ◽  
Artificial Intelligence Technology ◽  
One To One ◽  
Multi Classification

Artificial intelligence technology has been applied very well in big data analysis such as data classification. In this paper, the application of the support vector machine (SVM) method from machine learning in the problem of multi-classification was analyzed. In order to improve the classification performance, an improved one-to-one SVM multi-classification method was creatively designed by combining SVM with the K-nearest neighbor (KNN) method. Then the method was tested using UCI public data set, Statlog statistical data set and actual data. The results showed that the overall classification accuracy of the one-to-many SVM, one-to-one SVM and improved one-to-one SVM were 72.5%, 77.25% and 91.5% respectively in the classification of UCI publication data set and Statlog statistical data set, and the total classification accuracy of the neural network, decision tree, basic one-to-one SVM, directed acyclic graph improved one-to-one SVM and fuzzy decision method improved one-to-one SVM and improved one-to-one SVM proposed in this study was 83.98%, 84.55%, 74.07%, 81.5%, 82.68% and 92.9% respectively in the classification of fault data of transformer, which demonstrated the improved one-to-one SVM had good reliability. This study provides some theoretical bases for the application of methods such as machine learning in big data analysis.

scholarly journals POLYMORPHIC SBD PREPROCESSOR: A PREPROCESSING APPROACH FOR SOCIAL BIG DATA

2020 ◽  
Vol 11 (6) ◽  
pp. 953-961
Author(s):  
Amit K. Jadiya ◽  
Archana Chaudhary ◽  
Ramesh Thakur
Keyword(s):  
Big Data ◽  
Data Analysis ◽  
Missing Values ◽  
Data Preprocessing ◽  
Big Data Analysis ◽  
Analysis Approach ◽  
Product Reviews ◽  
Multiple Sources ◽  
Data Set ◽  
Social Big Data

In recent years, the social media has become a powerful tool for sharing people thoughts and feelings. As a result data is being generated, analyzed and used with a tremendous growth rate. The data generated by numerous updates, comments, news, opinions and product reviews in social websites is very useful for getting insights. As there are multiple sources, the size, speed and formats of the gathered data affects the overall quality of information. To achieve quality information, preprocessing step is very important and decides future roadmap for efficient big data analysis approach. In context to social big data we are addressing the preprocessing phase which includes cleaning of data, identifying noise, data normalization, data transformation, handling missing values and data integration. In this paper we have proposed a new approach polymorphic SBD (Social Big Data) preprocessor which provides efficient results with multiple social big data sets. Also available data preprocessing methods for big data are presented in this paper. After efficient and successful data preprocessing steps, the output data set will be efficient, well formed and suitable source for any big data analysis approach to be applied afterwards. The paper also presents an example case and evaluates min-max normalization, z-score normalization and data mapping for the case presented.

scholarly journals Self-Matching CAM: A Novel Accurate Visual Explanation of CNNs for SAR Image Interpretation

2021 ◽  
Vol 13 (9) ◽  
pp. 1772
Author(s):  
Zhenpeng Feng ◽  
Mingzhe Zhu ◽  
Ljubiša Stanković ◽  
Hongbing Ji
Keyword(s):  
Target Recognition ◽  
Image Interpretation ◽  
Automatic Target Recognition ◽  
Complex Data ◽  
Sar Image ◽  
Feature Maps ◽  
Sar Images ◽  
Recognition Mechanism ◽  
Imaging Processing ◽  
Activation Mapping

Synthetic aperture radar (SAR) image interpretation has long been an important but challenging task in SAR imaging processing. Generally, SAR image interpretation comprises complex procedures including filtering, feature extraction, image segmentation, and target recognition, which greatly reduce the efficiency of data processing. In an era of deep learning, numerous automatic target recognition methods have been proposed based on convolutional neural networks (CNNs) due to their strong capabilities for data abstraction and mining. In contrast to general methods, CNNs own an end-to-end structure where complex data preprocessing is not needed, thus the efficiency can be improved dramatically once a CNN is well trained. However, the recognition mechanism of a CNN is unclear, which hinders its application in many scenarios. In this paper, Self-Matching class activation mapping (CAM) is proposed to visualize what a CNN learns from SAR images to make a decision. Self-Matching CAM assigns a pixel-wise weight matrix to feature maps of different channels by matching them with the input SAR image. By using Self-Matching CAM, the detailed information of the target can be well preserved in an accurate visual explanation heatmap of a CNN for SAR image interpretation. Numerous experiments on a benchmark dataset (MSTAR) verify the validity of Self-Matching CAM.

Sign in / Sign up

Export Citation Format

Share Document