scholarly journals Construction of Apple Leaf Diseases Identification Networks Based on Xception Fused by SE Module

2021 ◽  
Vol 11 (10) ◽  
pp. 4614
Author(s):  
Xiaofei Chao ◽  
Xiao Hu ◽  
Jingze Feng ◽  
Zhao Zhang ◽  
Meili Wang ◽  
...  
Keyword(s):  
Disease Control ◽  
Classification Accuracy ◽  
Mobile Terminal ◽  
Disease Classification ◽  
Memory Usage ◽  
Compression Scheme ◽  
Accurate Identification ◽  
Average Classification Accuracy ◽  
Network Compression ◽  
Control And Management

The fast and accurate identification of apple leaf diseases is beneficial for disease control and management of apple orchards. An improved network for apple leaf disease classification and a lightweight model for mobile terminal usage was designed in this paper. First, we proposed SE-DEEP block to fuse the Squeeze-and-Excitation (SE) module with the Xception network to get the SE_Xception network, where the SE module is inserted between the depth-wise convolution and point-wise convolution of the depth-wise separable convolution layer. Therefore, the feature channels from the lower layers could be directly weighted, which made the model more sensitive to the principal features of the classification task. Second, we designed a lightweight network, named SE_miniXception, by reducing the depth and width of SE_Xception. Experimental results show that the average classification accuracy of SE_Xception is 99.40%, which is 1.99% higher than Xception. The average classification accuracy of SE_miniXception is 97.01%, which is 1.60% and 1.22% higher than MobileNetV1 and ShuffleNet, respectively, while its number of parameters is less than those of MobileNet and ShuffleNet. The minimized network decreases the memory usage and FLOPs, and accelerates the recognition speed from 15 to 7 milliseconds per image. Our proposed SE-DEEP block provides a choice for improving network accuracy and our network compression scheme provides ideas to lightweight existing networks.

scholarly journals Time–Frequency-Analysis-Based Blind Modulation Classification for Multiple-Antenna Systems

Sensors ◽  
2021 ◽  
Vol 21 (1) ◽  
pp. 231
Author(s):  
Weiheng Jiang ◽  
Xiaogang Wu ◽  
Yimou Wang ◽  
Bolin Chen ◽  
Wenjiang Feng ◽  
...  
Keyword(s):  
Frequency Analysis ◽  
Classification Accuracy ◽  
Communication Systems ◽  
Mimo Systems ◽  
Decision Fusion ◽  
Frequency Characteristics ◽  
Modulation Classification ◽  
Time Frequency Analysis ◽  
Time Frequency ◽  
Average Classification Accuracy

Blind modulation classification is an important step in implementing cognitive radio networks. The multiple-input multiple-output (MIMO) technique is widely used in military and civil communication systems. Due to the lack of prior information about channel parameters and the overlapping of signals in MIMO systems, the traditional likelihood-based and feature-based approaches cannot be applied in these scenarios directly. Hence, in this paper, to resolve the problem of blind modulation classification in MIMO systems, the time–frequency analysis method based on the windowed short-time Fourier transform was used to analyze the time–frequency characteristics of time-domain modulated signals. Then, the extracted time–frequency characteristics are converted into red–green–blue (RGB) spectrogram images, and the convolutional neural network based on transfer learning was applied to classify the modulation types according to the RGB spectrogram images. Finally, a decision fusion module was used to fuse the classification results of all the receiving antennas. Through simulations, we analyzed the classification performance at different signal-to-noise ratios (SNRs); the results indicate that, for the single-input single-output (SISO) network, our proposed scheme can achieve 92.37% and 99.12% average classification accuracy at SNRs of −4 and 10 dB, respectively. For the MIMO network, our scheme achieves 80.42% and 87.92% average classification accuracy at −4 and 10 dB, respectively. The proposed method greatly improves the accuracy of modulation classification in MIMO networks.

scholarly journals Analytical Comparison of Two Emotion Classification Models Based on Convolutional Neural Networks

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Huiping Jiang ◽  
Demeng Wu ◽  
Rui Jiao ◽  
Zongnan Wang
Keyword(s):  
Wavelet Transform ◽  
Classification Accuracy ◽  
Eeg Signal ◽  
Classification Models ◽  
Emotion Classification ◽  
Common Space ◽  
Average Classification Accuracy ◽  
The Common ◽  
The Brain

Electroencephalography (EEG) is the measurement of neuronal activity in different areas of the brain through the use of electrodes. As EEG signal technology has matured over the years, it has been applied in various methods to EEG emotion recognition, most significantly including the use of convolutional neural network (CNN). However, these methods are still not ideal, and shortcomings have been found in the results of some models of EEG feature extraction and classification. In this study, two CNN models were selected for the extraction and classification of preprocessed data, namely, common spatial patterns- (CSP-) CNN and wavelet transform- (WT-) CNN. Using the CSP-CNN, we first used the common space model to reduce dimensionality and then applied the CNN directly to extract and classify the features of the EEG; while, with the WT-CNN model, we used the wavelet transform to extract EEG features, thereafter applying the CNN for classification. The EEG classification results of these two classification models were subsequently analyzed and compared, with the average classification accuracy of the CSP-CNN model found to be 80.56%, and the average classification accuracy of the WT-CNN model measured to 86.90%. Thus, the findings of this study show that the average classification accuracy of the WT-CNN model was 6.34% higher than that of the CSP-CNN.

scholarly journals Identification of Potential Biomarkers using Improved Ranked Guided Iterative Feature Elimination

2021 ◽  
Vol 11 (1) ◽  
pp. 35-43
Author(s):  
Wen Xin Ng ◽  
Weng Howe Chan
Keyword(s):  
Gene Expression ◽  
Classification Accuracy ◽  
Disease Classification ◽  
High Dimensional ◽  
Feature Subset ◽  
Relevant Feature ◽  
Selection Methods ◽  
Additional Filter ◽  
Potential Biomarkers ◽  
Biological Context

In healthcare, biomarkers serve an important role in disease classification. Many existing works are focusing in identifying potential biomarkers from gene expression. Moreover, the large number of redundant features in a high dimensional dataset such as gene expression would introduce bias in the classifier and reduce the classifier’s performance. Embedded feature selection methods such as ranked guided iterative feature elimination have been widely adopted owing to the good performance in identification of informative features. However, method like ranked guided iterative feature elimination does not consider the redundancy of the features. Thus, this paper proposes an improved ranked guided iterative feature elimination method by introducing an additional filter selection based on minimum redundancy maximum relevance to filter out redundant features and maintain the relevant feature subset to be ranked and used for classification. Experiments are done using two gene expression datasets for prostate cancer and central nervous system. The performance of the classification is measured in terms of accuracy and compared with existing methods. Meanwhile, biological context verification of the identified features is done through available knowledge databases. Our method shows improved classification accuracy, and the selected genes were found to have relationship with the diseases.

scholarly journals Design and Implementation of an Efficient Rose Leaf Disease Detection using K-Nearest Neighbours

2020 ◽  
Vol 9 (3) ◽  
pp. 21-27
Keyword(s):  
Plant Disease ◽  
Spot Size ◽  
Extraction Process ◽  
Disease Classification ◽  
Identification System ◽  
Automated Identification ◽  
Accurate Identification ◽  
Leaf Disease ◽  
Key Factor ◽  
Agriculture Products

Plants are prone to different diseases caused by multiple reasons like environmental conditions, light, bacteria, and fungus. These diseases always have some physical characteristics on the leaves, stems, and fruit, such as changes in natural appearance, spot, size, etc. Due to similar patterns, distinguishing and identifying category of plant disease is the most challenging task. Therefore, efficient and flawless mechanisms should be discovered earlier so that accurate identification and prevention can be performed to avoid several losses of the entire plant. Therefore, an automated identification system can be a key factor in preventing loss in the cultivation and maintaining high quality of agriculture products. This paper introduces modeling of rose plant leaf disease classification technique using feature extraction process and supervised learning mechanism. The outcome of the proposed study justifies the scope of the proposed system in terms of accuracy towards the classification of different kind of rose plant disease.

scholarly journals 1D conditional generative adversarial network for spectrum-to-spectrum translation of simulated chemical reflectance signatures

2021 ◽  
Author(s):  
Cara Murphy ◽  
John Kerekes
Keyword(s):  
Classification Accuracy ◽  
Domain Adaptation ◽  
Real Data ◽  
Training Set ◽  
Generative Adversarial Network ◽  
Average Classification Accuracy ◽  
Adversarial Network ◽  
Chemical Residues ◽  
Reflectance Data

The classification of trace chemical residues through active spectroscopic sensing is challenging due to the lack of physics-based models that can accurately predict spectra. To overcome this challenge, we leveraged the field of domain adaptation to translate data from the simulated to the measured domain for training a classifier. We developed the first 1D conditional generative adversarial network (GAN) to perform spectrum-to-spectrum translation of reflectance signatures. We applied the 1D conditional GAN to a library of simulated spectra and quantified the improvement in classification accuracy on real data using the translated spectra for training the classifier. Using the GAN-translated library, the average classification accuracy increased from 0.622 to 0.723 on real chemical reflectance data, including data from chemicals not included in the GAN training set.

scholarly journals A Novel Hybrid Approach for Chronic Disease Classification

2020 ◽  
Vol 15 (1) ◽  
pp. 1-19 ◽  
Author(s):  
Divya Jain ◽  
Vijendra Singh
Keyword(s):  
Chronic Disease ◽  
Classification Accuracy ◽  
Hybrid Approach ◽  
Substantial Reduction ◽  
Disease Classification ◽  
Computational Time ◽  
Second Phase ◽  
Two Phase ◽  
Hybrid Classification ◽  
Diagnostic Framework

A two-phase diagnostic framework based on hybrid classification for the diagnosis of chronic disease is proposed. In the first phase, feature selection via ReliefF method and feature extraction via PCA method are incorporated. In the second phase, efficient optimization of SVM parameters via grid search method is performed. The proposed hybrid classification approach is then tested with seven popular chronic disease datasets using a cross-validation method. Experiments are then conducted to evaluate the presented classification method vis-à-vis four other existing classifiers that are applied on the same chronic disease datasets. Results show that the presented approach reduces approximately 40% of the extraneous and surplus features with substantial reduction in the execution time for mining all datasets, achieving the highest classification accuracy of 98.5%. It is concluded that with the presented approach, excellent classification accuracy is achieved for each chronic disease dataset while irrelevant and redundant features may be eliminated, thereby substantially reducing the diagnostic complexity and resulting computational time.

scholarly journals Improving Image-Based Plant Disease Classification With Generative Adversarial Network Under Limited Training Set

2020 ◽  
Vol 11 ◽  
Author(s):  
Luning Bi ◽  
Guiping Hu
Keyword(s):  
Classification Accuracy ◽  
Plant Disease ◽  
Data Augmentation ◽  
Plant Diseases ◽  
Disease Classification ◽  
Training Data ◽  
Training Dataset ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Overfitting Problem

Traditionally, plant disease recognition has mainly been done visually by human. It is often biased, time-consuming, and laborious. Machine learning methods based on plant leave images have been proposed to improve the disease recognition process. Convolutional neural networks (CNNs) have been adopted and proven to be very effective. Despite the good classification accuracy achieved by CNNs, the issue of limited training data remains. In most cases, the training dataset is often small due to significant effort in data collection and annotation. In this case, CNN methods tend to have the overfitting problem. In this paper, Wasserstein generative adversarial network with gradient penalty (WGAN-GP) is combined with label smoothing regularization (LSR) to improve the prediction accuracy and address the overfitting problem under limited training data. Experiments show that the proposed WGAN-GP enhanced classification method can improve the overall classification accuracy of plant diseases by 24.4% as compared to 20.2% using classic data augmentation and 22% using synthetic samples without LSR.

scholarly journals Continental-scale variation in otolith geochemistry of juvenile American shad (Alosa sapidissima)

2008 ◽  
Vol 65 (12) ◽  
pp. 2623-2635 ◽  
Author(s):  
Benjamin D. Walther ◽  
Simon R. Thorrold
Keyword(s):  
Water Chemistry ◽  
Classification Accuracy ◽  
High Spatial Resolution ◽  
American Shad ◽  
Native Range ◽  
Large Spatial Scale ◽  
Alosa Sapidissima ◽  
Continental Scale ◽  
Average Classification Accuracy ◽  
Annual Cohort

We assembled a comprehensive atlas of geochemical signatures in juvenile American shad ( Alosa sapidissima ) to discriminate natal river origins on a large spatial scale and at a high spatial resolution. Otoliths and (or) water samples were collected from 20 major spawning rivers from Florida to Quebec and were analyzed for elemental (Mg:Ca, Mn:Ca, Sr:Ca, and Ba:Ca) and isotope (87Sr:86Sr and δ18O) ratios. We examined correlations between water chemistry and otolith composition for five rivers where both were sampled. While Sr:Ca, Ba:Ca, 87Sr:86Sr, and δ18O values in otoliths reflected those ratios in ambient waters, Mg:Ca and Mn:Ca ratios in otoliths varied independently of water chemistry. Geochemical signatures were highly distinct among rivers, with an average classification accuracy of 93% using only those variables where otolith values were accurately predicted from water chemistry data. The study represents the largest assembled database of otolith signatures from the entire native range of a species, encompassing approximately 2700 km of coastline and 19 degrees of latitude and including all major extant spawning populations. This database will allow reliable estimates of natal origins of migrating ocean-phase American shad from the 2004 annual cohort in the future.

Image Texture Based Hybrid Diagnostic Tool for Kidney Disease Classification

2018 ◽  
Vol 8 (9) ◽  
pp. 1899-1908
Author(s):  
P. Sreelatha ◽  
M. Ezhilarasi
Keyword(s):  
Diagnostic Tool ◽  
Classification Accuracy ◽  
Wavelet Coefficient ◽  
Region Of Interest ◽  
Principal Component ◽  
Disease Classification ◽  
Image Texture ◽  
Support Vector ◽  
Adaptive Wavelet ◽  
Medical Diagnostic

The identification of chronic medical conditions and its associated mortality has led to the emergence of less invasive methods for medical diagnostic imaging. This work proposes a Computer Aided Diagnostic tool useful in automatic classification of kidney images as normal, simple cysts, kidney stones and the less investigated complex cystic renal cell carcinoma. The first part of the work investigates an effective despeckling algorithm with a proposed adaptive wavelet based denoising technique. Encouraging increased PSNR values ranging from 15 dB to 24 dB were obtained. Second part of work suggests a set of wavelet coefficient based feature set which showed a classification accuracy of 92.2%, better by 20.3% to 0.8% against existing methods. The final part of the work to develop a complete tool for kidney image classification combines the proposed wavelet based features with three existing statistical based feature sets yielded a classification accuracy of 96.9%. The suggested features were extracted from the region of interest from an image set. A reduced feature set of 18 from the original size of 163 was obtained using principal component analysis and applied for training a support vector machine classifier.

Sign in / Sign up

Export Citation Format

Share Document