Showing papers by "Rong Qu published in 2017"

Saikosaponin D relieves unpredictable chronic mild stress induced depressive-like behavior in rats: involvement of HPA axis and hippocampal neurogenesis.

Abstract: Saikosaponin D (SSD), a major bioactive component isolated from Radix Bupleuri, has been reported to exert neuroprotective properties. The present study was designed to investigate the anti-depressant-like effects and the potential mechanisms of SSD. Behavioural tests including sucrose preference test (SPT), open field test (OFT) and forced swim test (FST) were performed to study the antidepressant-like effects of SSD. In addition, we examined corticosterone and glucocorticoid receptor (GR) levels to evaluate hypothalamic-pituitary-adrenal (HPA) axis function. Furthermore, hippocampal neurogenesis was assessed by testing doublecortin (DCX) levels, and neurotrophic molecule levels were also investigated in the hippocampus of rats. We found that unpredictable chronic mild stress (UCMS) rats displayed lost body weight, decreased sucrose consumption in SPT, reduced locomotive activity in OFT, and increased immobility time in FST. Chronic treatment with SSD (0.75, 1.50 mg/kg) remarkably ameliorated the behavioral deficiency induced by UCMS procedure. SSD administration downregulated elevated serum corticosterone levels, as well as alleviated the suppression of GR expression and nuclear translocation caused by UCMS, suggesting that SSD is able to remit the dysfunction of HPA axis. In addition, Western blot and immunohistochemistry analysis showed that SSD treatment significantly increased the generation of neurons in the hippocampus of UCMS rats indicated by elevated DCX levels. Moreover, hippocampal neurotrophic molecule levels of UCMS rats such as phosphorylated cAMP response element binding protein (p-CREB) and brain-derived neurotrophic factor (BDNF) were raised after SSD treatment. Together, Our results suggest that SSD opposed UCMS-induced depressive behaviors in rats, which was mediated, partially, by the enhancement of HPA axis function and consolidation of hippocampal neurogenesis.

Survey on cyber security of CAV

Abstract: With the ever fast developments of technologies in science and engineering, it is believed that CAV (connected and autonomous vehicles) will come into our daily life soon CAV could be used in many different aspects in our lives such as public transportation and agriculture, and so on Although CAV will bring huge benefits to our lives and society, issues such as cyber security threats, which may reveal drivers' private information or even pose threat to driver's life, present significant challenges before CAV can be utilised in our society In computer science, there is a clear category of cyber security attacks while there is no specific survey on cyber security of CAV This paper overviews different passive and active cyber security attacks which may be faced by CAV We also present solutions of each of these attacks based on the current state-of-the-art, and discuss future improvements in research on CAV cyber security

CNN and selective attention mechanism based SAR image target detection method

Abstract: The invention discloses a CNN and selective attention mechanism based SAR image target detection method. An SAR image is obtained; a training data set is expanded; a classification model composed of the CNN is constructed; the expanded training data set is used to train the classification model; significance test is carried out on a test image via a simple attention model (a spectral residual error method) of image visual significance to obtain a significant characteristic image; and morphological processing is carried out on the significant characteristic image, the processed characteristic image is marked with connected domains, target candidate areas corresponding to different mass centers are extracted by taking the mass centers of the connected domains as the centers, and the target candidate areas are translated within pixels in the surrounding to generate an target detection result. According to the invention, the CNN and the selective attention mechanism are applied to SAR image target detection in a combined way, the efficiency and accuracy of SAR image target detection are improved, the method can be applied to target classification and identification, and the problem that detection in the prior art is low in detection efficiency and accuracy is solved mainly.

Change detection in SAR images based on the salient map guidance and an accelerated genetic algorithm

Abstract: This paper proposes a change detection algorithm in synthetic aperture radar (SAR) images based on the salient image guidance and an accelerated genetic algorithm (S-aGA). The difference image is first generated by logarithm ratio operator based on the bi-temporal SAR images acquired in the same region. Then a saliency detection model is applied in the difference image to extract the salient regions containing the changed class pixels. The salient regions are further divided by fuzzy c-means (FCM) clustering algorithm into three categories: changed class (set of pixels with high gray values), unchanged class (set of pixels with low gray values) and undetermined class (set of pixels with middle gray value, which are difficult to classify). Finally, the proposed accelerated GA is applied to explore the reduced search space formed by the undetermined-class pixels according to an objective function considering neighborhood information. In S-aGA, an efficient mutation operator is designed by using the neighborhood information of undetermined-class pixels as the heuristic information to determine the mutation probability of each undetermined-class pixel adaptively, which accelerates the convergence of the GA significantly. The experimental results on two data sets demonstrate the efficiency of the proposed S-aGA. On the whole, S-aGA outperforms five other existing methods including the simple GA in terms of detection accuracy. In addition, S-aGA could obtain satisfying solution within limited generations, converging much faster than the simple GA.

SAR image target detection method based on full convolutional neural network

Abstract: The invention discloses an SAR image target detection method based on a full convolutional neural network. The method mainly solves the problem of low accuracy and slow detection speed in the prior art, and is characterized by obtaining an SAR image; expanding a training dataset; constructing a nine-layer full convolutional neural network; training the full convolutional neural network through the expanded training dataset; inputting a test image into a trained model for significance test to obtain an output significance feature graph; carrying out morphological processing on the significance feature graph; carrying out connected domain labeling on the processed feature graph; with the mass center of each connected domain being the center, extracting a detection slice corresponding to each target mass center; and labeling each detection slice in the input original SAR image and obtaining a target test result of the test data. The full convolutional neural network is applied to SAR image target detection, thereby improving SAR image target detection speed and accuracy; and the method can also be used for object identification.

High-resolution SAR image classification method based on non-down-sampling contourlet full-convolution network

Abstract: A high-resolution SAR image classification method based on a non-down-sampling contourlet full-convolution network is provided, which comprises: inputting a high-resolution SAR image to be classified; performing multi-layer non-down-sampling contourlet transform on each pixel in the image; obtaining the low-frequency coefficient and the high-frequency coefficient of each pixel; selecting and fusing the low-frequency coefficients and high-frequency coefficients to form a pixel-based characteristic matrix F; normalizing the element values in the characteristic matrix F to obtain a normalized characteristic matrix F1; dicing the normalized characteristic matrix F1 to obtain a characteristic block matrix F2 used as sample data; constructing a training data set characteristic matrix W1 and a testing data set characteristic matrix W2; constructing a classification model based on a full convolution neural network; training the classification model; utilizing the well-trained model to classify the testing data set T to obtain the category of each pixel in the testing data set T; comparing the obtained category of each pixel with a class diagram; and calculating the classification accuracy. With the method, the classification accuracy and speed are increased.

Multispectral remote sensing image terrain classification method based on deep and semi-supervised transfer learning

Abstract: The invention discloses a multispectral remote sensing image terrain classification method based on deep and semi-supervised transfer learning. A training data set and kNN data are extracted according to ground truth; the training data set is divided into two parts to be trained respectively; a multispectral image to be classified is inputted, and two classification result images are obtained from two CNN models; two kNN nearest neighbor algorithm images are constructed according to the training samples; the tested data are extracted by using the two classification result images, and the data are classified by using the kNN nearest neighbor algorithm; the classification result images are updated; the training samples and the kNN training samples of cooperative training are updated; and two cooperative training CNN networks are trained again, and the points having the class label of the test data set are classified by using the trained model so that the class of partial pixel points in the test data set is obtained and compared with the real class label. The k nearest neighbor algorithm and the sample similarity are introduced so that deviation of cooperative training can be prevented, the classification accuracy in case of insufficient training samples can be enhanced and thus the method can be used for target recognition.

Multi-spectral image classification method based on threshold self-adaption and convolutional neural network

Abstract: The invention discloses a multi-spectral image classification method based on threshold self-adaption and a convolutional neural network. The method comprises the following steps of inputting multispectral images of different time phases and different wave bands of satellites to be classified, and carrying out normalization on all pixels of a marked portion of a same wave band image in all cities; stacking selected 9 wave bands into one image and taking the image as a training data set; constructing a classification model based on the convolutional neural network, using the training data set to train the classification model so as to acquire a probability model based on OSM, using the model and a confidence coefficient strategy to adjust a softmax output result and acquiring a final classification model, and finally, uploading a test result to an IEEE website so as to acquire classification accuracy. By using the multi-spectral image classification method, characteristics that there are a lot of wave bands, a data volume is large and there are a lot of information redundancy in the multispectral images are fully used so that a problem that surface features of complex types are difficult to classify is solved, classification accuracy is increased, an error dividing rate is reduced and a classification speed can be increased.

Polarization SAR image classification method based on complex contour wave convolution neural network

Abstract: The invention discloses a polarization SAR image classification method based on a complex contour wave convolution neural network, and a problem of low classification accuracy in the prior art is mainly solved. The method comprises the steps of (1) inputting and normalizing a polarization coherent matrix T of a polarization SAR image to be classified, (2) according to the normalized matrix, constructing characteristic matrixes of a training data set and a test data set, (3) constructing a complex convolution neural network, and thus obtaining a complex contour wave convolution neural network, (4) training the complex contour wave convolution neural network by using the training data set, and obtaining a trained model, and (5) inputting the characteristic matrix of a test data set into the trained model to carry out classification, and obtaining a classification result. According to the method, the convolution neural network is extended to a complex domain to carry out operation and extract image characteristics of multiple scales, multiple directions and multiple resolution characteristics, the classification precision of the polarization SAR image is effectively improved, and the method can be used for target detection and identification.

Remote sensing image ground object classification method based on superpixel coding and convolution neural network

Abstract: The invention discloses a remote sensing image ground object classification method based on superpixel coding and convolution neural network, using adaptive superpixel coding and double channel convolution neural network. The remote sensing image ground object classification method based on superpixel coding and convolution neural network includes the steps: utilizing a superpixel algorithm to perform image pre-segmentation; using a cluster method to merge neighboring and similar superpixel blocks, setting the size of the taken blocks, constructing three double channel convolution neural networks with different input size; inputting samples with different taken block size into the corresponding network; using the convolution neural networks to extract the data characteristics of two sensors respectively; merging the extracted characteristics for classification; and according to the size of the merged pixel block, determining the size of the taken blocks of the samples, and realizing adaptive selection of the utilized neighborhood information. The remote sensing image ground object classification method based on superpixel coding and convolution neural network can realize adaptive selection of the utilized neighborhood information to enable the neighborhood information to realize positive feedback effect and preferably utilize the neighborhood information to send the samples to different networks according to the neighborhood information so as to enable the samples with similar distribution to enter the same network, thus effectively improving the classification accuracy.

Polarized SAR land feature classification method based on full convolution neural network

Abstract: The invention discloses a polarized SAR land feature classification method based on a full convolution neural network, comprising: performing Pauli decomposition on a to-be-classified polarized scattering matrix S to obtain the odd scattering coefficient, the even scattering coefficient and the volume scattering coefficient; using the odd scattering coefficient, the even scatter coefficient and the volume scattering coefficient as the three-dimensional image characteristics F of the polarized SAR image; then converting the obtained three-dimensional image characteristics matrix F into an RGB image F1; randomly selecting m x n pixel blocks in the RGB image F1 as training samples; using the whole RGB image F1 as a testing sample; re-constructing a full convolution neural network model; training the training samples through the full convolution neural network to obtain a trained model; and then, through the trained model, classifying the test set and obtaining the classification result The method of the invention can solve the problem of low time efficiency in the prior art and shorten the running time under the condition of high classification accuracy

Polarimetric SAR (synthetic aperture radar) image target detection method based on multipolarization features and FCN (full convolutional)-CRF (conditional random field) fusion network

Abstract: The invention discloses a polarimetric SAR (synthetic aperture radar) image target detection method based on multipolarization features and an FCN (fully convolutional network)-CRF (conditional random field) fusion network. The invention aims to solve the problem of low detection accuracy of a polarimetric SAR artificial target of the prior art. The method includes the following steps that: a polarized SAR image to be detected is inputted, and Lee filtering is performed on the polarization coherent matrix T of the polarimetric SAR image; Pauli decomposition is performed on a polarimetric scattering matrix S, so that a pixel point-based feature matrix F1 can be formed; and Yamaguchi decomposition is performed on the filtered coherent matrix T, so that a pixel-based feature matrix F2 can be formed. According to the method of the invention, the multi-polarization feature and the FCN-CRF-based fusion network are applied to the detection of a polarimetric SAR artificial target, and therefore, the detection accuracy of the polarimetric SAR artificial target can be improved; and the method can be applied to multi-target classification tasks.

A multispectral image classification method based on a double-channel multi-feature fusion network

Abstract: The invention provides a multispectral image classification method based on a double-channel multi-feature fusion network. The method comprises the steps of: fusing multispectral features of different wave bands of two satellites to obtain features L and L'; performing normalization processing on the L and the L' to obtain Lnorm and L' norm; selecting pixel blocks randomly on the Lnorm and the L' norm to form a training set and a validation set and form feature matrixes Wtrain and Wval based on image blocks and obtaining a feature matrix Wtest of a testing set according to a feature matrix of sao-paulo city; building a classification model of a double-channel all-convolutional neural network; training the classification model by using the feature matrix Wtrain of the training data set and the feature matrix Wval of the validation data set; classifying the feature matrix Wtest of the test data set by using the trained classification model. According to the invention, the double-channel all-convolutional neural network is used for multispectral image classification; compared with a common all-convolutional neural network, the double-channel all-convolutional neural network can increase the classification accuracy.

Polarity SAR target detection method based on FCN-CRF master-slave network

Abstract: The invention provides a polarity SAR target detection method based on a FCN-CRF master-slave network. The method comprises steps of inputting a to-be-detected polarity SAR image, and carrying out delicate polarity Lee filtering on a polarity coherence matrix T of the polarity SAR image to filter coherent noise so as to obtain the filtered coherence matrix, wherein each element of the filtered coherence matrix is a 3*3 matrix, that is to say, each pixel point has nine-dimensional features. According to the invention, by expanding image block features into pixel-level features, the correlation degree of selected training samples through matching of pixel points of a region of interest is quite high and quite effective; the feature image blocks with the pixel points of the region of interest whose quantity is less than 50% of the whole image block will not participate in following calculation, so the operand is greatly reduced and the detection efficiency is improved; by using the Lee filtering to pre-process of the original polarity SAR image, coherence spot noise is effectively reduced and image quality and detection performance are improved; and by use of spiral scattering components corresponding to urban buildings obtained through the Yamaguchi decomposition, features of polarity SAR artificial targets are effectively extracted, and detection precision of the artificial targets is improved.

Multi-source remote sensing image surface object classification method based on double-channel convolution step network

Abstract: The invention discloses a multi-source remote sensing image surface object classification method based on a double-channel convolution step network. The multispectral data of regions to be classified obtained by a landsat-8 sensor and a sentinel-2 sensor are normalized by suing ENVI software so as to obtain the normalized multispectral data; 28x28 blocks around each element of the normalized multispectral data are taken to represent the original element value so as to form a feature matrix based on the image blocks; multiple blocks are randomly selected from each class to for training data sets L and S; a multi-source remote sensing image surface object classification model based on the double-channel convolution step network is constructed; the multi-source remote sensing image surface object classification model based on the double-channel convolution step network is trained by using the training data sets L and S; and test data sets are classified by using the trained multi-source remote sensing image surface object classification model based on the double-channel convolution step network. The high multi-source image classification accuracy can be acquired by only using less class tag samples so that the method can be used for target detection.

Polarimetric SAR image classification method based on DCGAN

Abstract: The invention discloses a polarimetric SAR image classification method based on DCGAN. The method comprises the following steps: 1) obtaining an odd-order scattering coefficient, an even-order scattering coefficient, and a volume scattering coefficient, establishing a characteristic matrix F based on pixel points; 2) normalizing each element value in the characteristic matrix F based on pixel points to [0,1], and calling a result of normalization as a feature matrix F1; 3) replacing each element in the feature matrix F1 by 64x64 image blocks around each elements, to obtain a feature matrix F2 based on the image blocks; 4) establishing a feature matrix W1 of a no-label training dataset D1 and a feature matrix W2 of a labeled training dataset D2; 5) establishing a feature matrix W3 of a superpixel clustering center of a testing dataset T; 6) obtaining a trained training network model DCGAN; 7) establishing a determining classification network model, and then through the determining classification network model, classifying the feature matrix W3. The method can realize classification of a polarimetric SAR image, and classification precision is relatively high.

Non-subsample contourlet DCGAN-adopted polarized SAR image classification method

Abstract: The invention discloses a non-subsample contourlet DCGAN-based polarized SAR image classification method. The method comprises the following steps of: inputting a to-be-classified polarized SAR image to carry out Pauli decomposition; forming an image block-based data set by 32*32 blocks by using a normalized dataset; constructing a no-label training dataset, a label training dataset and a test dataset, dividing superpixel blocks for the Pauli decomposed pseudo color graph by utilizing an SLIC superpixel algorithm, constructing a non-subsample contourlet DCGAN, and training the non-subsample contourlet DCGAN of a training network model by using the no-label training dataset; inputting the label training dataset into a discrimination and classification network model to train a softmax classifier, and finely adjusting parameters of the whole discrimination and classification network; and classifying a superpixel clustering center of the test dataset by utilizing the trained discrimination and classification network model, and marking the category of each pixel point in the test dataset. According to the method, the polarized SAR image classification precision can be improved, so that the method can be used for target identification, tracking and positioning.

SAR image target classification method based on NSCT double CNN channels and selective attention mechanism

Abstract: The invention discloses an SAR image target classification method based on NSCT double CNN channels and a selective attention mechanism The method comprises steps that training sample sets D1 and D2 for target detection and classification are acquired; the D1 and D2 are expanded to acquire sample sets D3 and D4; models M1 and M2 for target detection and classification are respectively trained; significance detection and morphological processing on test images are carried out, connected domain marking is further carried out, target candidate areas corresponding to a connected domain mass center are extracted, translation in multiple surrounding pixel points is carried out, and the target candidate areas are generated; classification determination of the target candidate areas is carried out through utilizing the M1, and accurate positioning of a target is acquired; a final class of the target is determined through voting decision after M2 classification The method is advantaged in that a non-down-sampling contour wave layer is added, low frequency and high frequency characteristic images are inputted to a double channel CNN to form the NSCT double channel CNN, the selective attention mechanism is applied to SAR image classification, SAR image target detection classification accuracy is improved, and a problem of low target classification accuracy in the prior art is solved

Information core optimization using Evolutionary Algorithm with Elite Population in recommender systems

Abstract: Recommender system (RS) plays an important role in helping users find the information they are interested in and providing accurate personality recommendation. It has been found that among all the users, there are some user groups called “core users” or “information core” whose historical behavior data are more reliable, objective and positive for making recommendations. Finding the information core is of great interests to greatly increase the speed of online recommendation. There is no general method to identify core users in the existing literatures. In this paper, a general method of finding information core is proposed by modelling this problem as a combinatorial optimization problem. A novel Evolutionary Algorithm with Elite Population (EA-EP) is presented to search for the information core, where an elite population with a new crossover mechanism named as ordered crossover is used to accelerate the evolution. Experiments are conducted on Movielens (100k) to validate the effectiveness of our proposed algorithm. Results show that EA-EP is able to effectively identify core users and leads to better recommendation accuracy compared to several existing greedy methods and the conventional collaborative filter (CF). In addition, EA-EP is shown to significantly reduce the time of online recommendation.

Polarized SAR image target detection method based on deep stairway network

Abstract: The invention discloses a polarized SAR image target detection method based on a deep stairway network, comprising the following steps: inputting a polarized SAR image to be detected and performing Lee filtering on the polarized coherent matrix T; solving the filtered T to obtain a polarized covariance matrix C; performing Yamahachi decomposition on the polarized covariance matrix C to form a pixel-based characteristics matrix F; normalizing the F and extracting blocks for each element of the normalized characteristics matrix F1 and listing them in a column; forming a characteristics matrix F2 based on the image blocks; obtaining a training set D according to F2; using the SLIC algorithm in the super pixels to obtain a test set T; constructing a target detection model based on a deep stairway network; using the training data set D to train the target detection model; and utilizing the trained target detection model to classify the test data set T. According to the invention, a deep stairway network is adopted, which only uses a small amount of labeled samples to obtain high target detection precision. The present invention can be used for the classification of ground targets.

SAE and saliency detection-based high-resolution SAR image change detection method

Abstract: The invention provides a SAE and saliency detection-based high-resolution SAR image change detection method. According to the method, different sizes of image blocks are extracted from two registered SAR images of the same region and different time phases, and the extracted image blocks are adopted as a first training data set and a second training data set, respectively. The two training data sets are normalized to [ 0, 1 ], respectively. After that, two self-coding networks of a three-layer stacked structure are respectively constructed, and the characteristic number of each layer of the networks is determined. Meanwhile, the weight and the bias are randomly initialized. The two normalized training data sets are respectively fed into the two self-coding networks of the three-layer stacked structure, and the weight and the bias of each layer are obtained through training. Two images are fed into the well trained networks respectively and then the features of the two images are obtained. The difference between the two images is obtained in a feature domain, and a threshold segmentation difference chart is determined through the thresholding method of the above difference. In this way, salient regions are obtained respectively. A final salient region is obtained through combining the two salient regions. Meanwhile, a final change detection result is obtained through the clustering algorithm. Therefore, the detection accuracy is effectively improved.

High resolution SAR image classification method based on deep convolutional step network

Abstract: The invention discloses a high resolution SAR image classification method based on the deep convolutional step network To-be-classified high resolution SAR images and the mark information thereof are inputted; a training data set D1 and a test data set D2 are constructed; normalization of characteristics of the data sets D1 and D2 is respectively carried out to acquire data sets D3 and D4; a classifier model based on the deep convolutional step network is constructed; the network is trained through utilizing the training data set D3; the test data set D4 is classified through utilizing the trained classification model The method is advantaged in that a few of labeled training samples can be fully utilized, the convolutional layer is further employed to effectively extract high-layer discrimination characteristics, and thereby relatively high classification precision can be realized

Pauli decomposition and depth residual network-based polarimetric SAR image classification method

Abstract: The invention discloses a Pauli decomposition and depth residual network-based polarimetric SAR image classification method. The method mainly solves the problems in the prior art that the classification accuracy is low and the neural network cannot be increased more deeply. According to the technical scheme of the invention, the method comprises the steps of inputting a to-be-classified polarimetric SAR image, subjecting a polarization scattering matrix S to Pauli decomposition, and forming a pixel-based feature matrix F; representing an original element value by 28*28 blocks around each element in the feature matrix F and forming an image block-based feature matrix; constructing a training data set D; subjecting the image to super-pixel treatment after the Pauli decomposition treatment and forming a data set T1; constructing a classification model based on a depth residual network; training the classification model by using the training data set so as to obtain a well trained model; inputting the data set T1 into the well trained model to classify the data set T1 and then obtaining a predictive label matrix T2 of the entire image; removing the pixels of the training data set from the matrix T2 and then calculating the accuracy. The method of the invention adopts the depth residual network, so that network layers are increased. Meanwhile, the image is processed by adopting super pixels, so that the features of the image are effectively learnt. The classification accuracy of polarimetric SAR images is improved. The method can be used for target recognition.

Variable-depth adaptive large meighbourhood search algorithm for Open Periodic Vehicle Routing Problem with time windows

Abstract: The Open Periodic Vehicle Routing Problem with Time Windows (OPVRPTW) is a practical transportation routing and scheduling problem arising from real-world scenarios. It shares some common features with some classic VRP variants. The problem has a tightly constrained large-scale solution space and requires well balanced diversification and intensification in search. In Variable Depth Neighbourhood Search, large neighbourhood depth prevents the search from trapping into local optima prematurely, while small depth provides thorough exploitation in local areas. Considering the multi-dimensional solution structure and tight constraints in OPVRPTW, a Variable-Depth Adaptive Large Neighbourhood Search (VD-ALNS) algorithm is proposed in this paper. Contributions of four tailored destroy operators and three repair operators at variable depths are investigated. Comparing to existing methods, VD-ALNS makes a good trade-off between exploration and exploitation, and produces promising results on both small and large size benchmark instances.

Solving the randomly generated university examination timetabling problem through Domain Transformation Approach (DTA)

Abstract: Amongst the wide-ranging areas of the timetabling problems, educational timetabling was reported as one of the most studied and researched areas in the timetabling literature. In this paper, our focus is the university examination timetabling. Despite many approaches proposed in the timetabling literature, it has been observed that there is no single heuristic that is able to solve a broad spectrum of scheduling problems because of the incorporation of problem-specific features in the heuristics. This observation calls for more extensive research and study into how to generate good quality schedules consistently. In order to solve the university examination timetabling problem systematically and efficiently, in our previous work, we have proposed an approach that we called a Domain Transformation Approach (DTA) which is underpinned by the insights from Granular Computing concept. We have tested DTA on some benchmark examination timetabling datasets, and the results obtained were very encouraging. Motivated by the previous encouraging results obtained, in this paper we will be analyzing the proposed method in different aspects. The objectives of this study include (1) To test the generality/applicability/universality of the proposed method (2) To compare and analyze the quality of the schedules generated by utilizing Hill Climbing (HC) optimization versus Genetic Algorithm (GA) optimization on a randomly generated benchmark. Based on the results obtained in this study, it was shown that our proposed DTA method has produced very encouraging results on randomly generated problems. Having said this, it was also shown that our proposed DTA method is very universal and applicable to different sets of examination timetabling problems.

High resolution SAR image change detection based on SPP Net and region of interest detection

Abstract: The invention discloses a high resolution SAR image change detection method based on the SPP Net and region of interest detection. The method comprises the steps of 1) obtaining two registered SAR images with different phases in the same region, then selecting a plurality of labeled data from the two registered SAR images with different phases in the same region, and taking the selected labeled data as training samples; 2) normalizing the training samples to between [0,1], and recording the normalized result as a sample X; 3) obtaining a trained SPP Net region of interest detection network; 4) obtaining a final region of interest; 5) obtaining two SAR images I1 and I2 after the interest detection; and 6) performing change detection on the two SAR images I1 and I2 after the interest detection obtained in the step 5) through the GKI based on image blocks to obtain a final change detection result map. The method realizes the change detection of large-scale and high-resolution SAR images and has a high change detection precision.

High-resolution SAR image classification method based on depth ladder network

Abstract: The invention discloses a high-resolution SAR image classification method based on a depth ladder network, mainly to solve the problem that the high-resolution SAR image has few data with class identifiers and a network can not be trained effectively The method comprises steps: a to-be-classified high-resolution SAR image and identifier information thereof are inputted; a training data set D1 and a test data set D2 are constructed; features of the data sets D1 and D2 are normalized to obtain data sets D3 and D4; a classifier model based on the depth ladder network is constructed; the training data set D3 is used for training the network; and the well-trained classifier model is used for classifying the test data set D4 The few training samples with class identifiers can be made full use of, and the high classification precision can be acquired