[신간의 별자리x] 우리/미술, 그리고 ‘슬픔의 박물관’

Hyperspectral image (HSI) classification is a most challenging task in hyperspectral remote sensing field due to unique characteristics of HSI data. It consists of huge number of bands with strong ...

Analysis of various optimizers on deep convolutional neural network model in the application of hyperspectral remote sensing image classification

Recently, many convolutional neural network (CNN) methods have been designed for hyperspectral image (HSI) classification since CNNs are able to produce good representations of data, which greatly benefits from a huge number of parameters. However, solving such a high-dimensional optimization problem often requires a large number of training samples in order to avoid overfitting. In addition, it is a typical nonconvex problem affected by many local minima and flat regions. To address these problems, in this article, we introduce the naive Gabor networks or Gabor-Nets that, for the first time in the literature, design and learn CNN kernels strictly in the form of Gabor filters, aiming to reduce the number of involved parameters and constrain the solution space and, hence, improve the performances of CNNs. Specifically, we develop an innovative phase-induced Gabor kernel, which is trickily designed to perform the Gabor feature learning via a linear combination of local low-frequency and high-frequency components of data controlled by the kernel phase. With the phase-induced Gabor kernel, the proposed Gabor-Nets gains the ability to automatically adapt to the local harmonic characteristics of the HSI data and, thus, yields more representative harmonic features. Also, this kernel can fulfill the traditional complex-valued Gabor filtering in a real-valued manner, hence making Gabor-Nets easily perform in a usual CNN thread. We evaluated our newly developed Gabor-Nets on three well-known HSIs, suggesting that our proposed Gabor-Nets can significantly improve the performance of CNNs, particularly with a small training set.

Naive Gabor Networks for Hyperspectral Image Classification

Network intrusion detection (NID) is an important method for network system administrators to detect various security holes. The performance of traditional NID methods can be affected when unknown or new attacks are detected. Compared with other machine learning methods, the intrusion detection method based on convolutional neural network (CNN) can significantly improve the accuracy of classification, but the convergence speed and generalization ability of CNN are not ideal in model training process resulting in a low true rate and a high false alarm rate. To solve the above problems, this paper proposes a deep multi-scale convolutional neural network (DMCNN) for network intrusion detection. Different levels of features in a large number of high-dimensional unlabeled original data are extracted by different scales convolution kernel. And the learning rate of network structure is optimized by batch normalization method to obtain the optimal feature representation of the raw data. We use NSL-KDD dataset as the benchmark thus we can compare the performance of our proposed method with other existing works. This dataset includes two testing sets which are the first one is KDDTest+ while the second one is $$\text {KDDTest}^{-21}$$
 which is more difficult to be classified. The experimental results reveal that the AC and TPR are higher through our DMCNN model. Especially, in terms of DOS, the AC appropriately reaches to 98%. DMCNN has a high intrusion detection accuracy and a low false alarm rate, which overcomes the limitations of using the traditional detection methods and makes the new approach an attractive one for practical intrusion detection.

A Network Intrusion Detection Method Based on Deep Multi-scale Convolutional Neural Network

Hyperspectral images are images containing rich spectral and spatial information widely used in remote sensing applications. The development of deep learning techniques has had a significant impact on the classification of hyperspectral images. Different Convolutional Neural Network architectures have been used in many hyperspectral image analysis studies. However, the high dimensions of the hyperspectral images increased the computational complexity. For this reason, dimensionality reduction has been used in the preprocessing stage in many studies. Another difficulty encountered in hyperspectral image classification studies is the need to consider both spectral and spatial features. When deep spatial and spectral features are to be extracted, problems such as loss of gradient properties and degradation due to increased depth arise. In this study, the 3D convolutional neural network (CNN) based ResNet50 method is proposed to solve these problems encountered in hyperspectral studies and to extract sufficient spatial spectral properties from the network. Principal Component Analysis (PCA) was used to reduce spectral band excess. The proposed method has been applied to Pavia University and Salinas data sets. Overall accuracy, average accuracy and kappa values were used to measure the performance of the method. Calculated overall accuracy, average accuracy, and kappa values are 99.99% for the Pavia University data set, and while the overall accuracy and kappa values were 99.99% for the Salinas data set, the average accuracy value was 99.98%.

Classification of Hyperspectral Images Using 3D CNN Based ResNet50

Hyperspectral satellite imagery (HSI) is an advanced technology for object detection because it provides a large amount of information. Thus, the classification of HSIs is very complicated, so the methods of reducing spectral or spatial information generally degrade the quality of classification. In order to solve this problem and guarantee faster and more efficient processing, we propose a smart feature extraction (SFE) and classification by convolutional neural network (2D-CNN) method made up of two parts. The first consists in reducing spectral information by a probabilistic method based on the Softmax function. The second is classification by processing batches of data in the proposed CNN network. The method was tested on two public hyperspectral images (Indian Pines and SalinasA) to prove its effectiveness in increasing classification accuracy and reducing computing time.

https://ietresearch.onlinelibrary.wiley.com/doi/pdf/10.1049/iet-ipr.2019.1282

Smart feature extraction and classification of hyperspectral images based on convolutional neural networks

Hyperspectral Imaging is a technology representing a scene via a large number of spectral bands. By increasing the amount of satellite information, the classification becomes more and more difficult. In recent years, deep learning has shown its effectiveness in classification and the identification of objects, especially via the Convolutional Neural Networks. In this paper, we proposed a deep classification method based on the CNN networks and a clustering algorithm; Our framework is composed of three steps, which are: (1) Automatic selection of the number and positions of kernels; (2) Adaptive selection of patches size; (3) Modification at the pooling layer; The proposed method is performed on three hyperspectral datasets: SalinasA, Indian Pines, and Pavia University. The obtained results reveal that the proposed method provide competitive results to the other approaches.

Modified Convolutional Neural Network based on Adaptive Patch Extraction for Hyperspectral Image Classification

Image classification by the convolutional neural network (CNN) has shown its great performances in recent years, in several areas, such as image processing and pattern recognition. However, there is still some improvement to do. The main problem in CNN is the initialisation of the number and size of the filters, which can obviously change the results. In this study, the authors assign three major contributions, based on the CNN model; (i) adaptive selection of the number of filters, (ii) using an adaptive size of the windows and (iii) using an adaptive size of the filters. The tests results, applied to different hyperspectral datasets (SalinasA, Pavia University, and Indian Pines), have proven that this framework is able to improve the accuracy of the hyperspectral image classification.

https://ietresearch.onlinelibrary.wiley.com/doi/pdf/10.1049/iet-ipr.2018.5063

Hyperspectral imaging classification based on convolutional neural networks by adaptive sizes of windows and filters

Convolutional neural networks (CNN) can learn deep feature representation for hyperspectral imagery (HSI) interpretation and attain excellent accuracy of classification if we have many training sam...

Framework for Automatic Selection of Kernels based on Convolutional Neural Networks and CkMeans Clustering Algorithm

Hyperspectral Satellite Images (HSI) presents a very interesting technology for mapping, environmental protection, and security. HSI is very rich in spectral and spatial characteristics, which are non-linear and highly correlated which makes classification difficult. In this paper, we propose a new approach to the reduction and classification of HSI. This deep approach consisting of a dual Convolutional Neural Networks (DCNN), which aims to improve precision and computing time. This approach involves two main steps; the first is to extract the spectral data and reduce it by CNN until a single value representing the active pixel is displayed. The second consists in classifying the only remaining spatial band on CNN until the class of each pixel is obtained. The tests were applied to three different hyperspectral data sets and showed the effectiveness of the proposed method.

Maissa Hamouda

Papers

Smart feature extraction and classification of hyperspectral images based on convolutional neural networks

Modified Convolutional Neural Network based on Adaptive Patch Extraction for Hyperspectral Image Classification

Hyperspectral imaging classification based on convolutional neural networks by adaptive sizes of windows and filters

Framework for Automatic Selection of Kernels based on Convolutional Neural Networks and CkMeans Clustering Algorithm

Dual Convolutional Neural Networks for Hyperspectral Satellite Images Classification (DCNN-HSI)