Explore the vast range of titles available.

Plant leaf diseases significantly impact crop yield and quality, causing substantial economic loss and risking food security. Despite significant progress in the field of automated plant disease diagnosis, there are still several challenges that need to be addressed. Accurate classification of plant leaf diseases at an early stage is crucial for diagnosis and effective treatment of these plant diseases. As the agricultural industry faces growing challenges from plant diseases, quickly identifying these diseases in a field environment while considering the computational resource limitations is more important than ever. To overcome these challenges, this study proposed a lightweight and compact convolutional neural network model, HPDC-Net (Hybrid Plant Disease Classification Network). The network used a block architecture with three blocks termed as Depth-wise Separable Convolution Block (DSCB), Dual-Path Adaptive Pooling Block (DAPB), and Channel-Wise Attention Refinement Block (CARB). The model extracts a robust but limited number of features due to the use of depth-wise separable convolutions in DSCB, making it accurate but lightweight. The proposed model has been trained to classify potato and tomato leaf diseases on three datasets. The model achieves a high accuracy score > 99% on all three datasets while keeping GFLOPs limited to 0.06 and the number of parameters to 0.52 M (for 10 classes) and 0.17 M (for 03 classes), yielding 19.82 FPS on CPU and 408.25 FPS on GPU in our setup. The code for implementation of proposed model is available on GitHub: https://github.com/ZahidFarooqKhan/HPDC-Net .

The efficient separation of coal and gangue remains a critical challenge in modern coal mining, directly impacting energy efficiency, environmental protection, and sustainable development. Current machine vision-based sorting methods face significant challenges in dense scenes, where label rewriting problems severely affect model performance, particularly when coal and gangue are closely distributed in conveyor belt images. This paper introduces CGDet (Coal and Gangue Detection), a novel compact convolutional neural network that addresses these challenges through two key innovations. First, we proposed an Object Distribution Density Measurement (ODDM) method to quantitatively analyze the distribution density of coal and gangue, enabling optimal selection of input and feature map resolutions to mitigate label rewriting issues. Second, we developed a Relative Resolution Object Scale Measurement (RROSM) method to assess object scales, guiding the design of a streamlined feature fusion structure that eliminates redundant components while maintaining detection accuracy. Experimental results demonstrate the effectiveness of our approach; CGDet achieved superior performance with AP50 and AR50 scores of 96.7% and 99.2% respectively, while reducing model parameters by 46.76%, computational cost by 47.94%, and inference time by 31.50% compared to traditional models. These improvements make CGDet particularly suitable for real-time coal and gangue sorting in underground mining environments, where computational resources are limited but high accuracy is essential. Our work provides a new perspective on designing compact yet high-performance object detection networks for dense scene applications.

In recent years, the advancement of pattern recognition algorithms, specifically the deep learning-related techniques, have propelled a tremendous amount of researches in fall event recognition systems. It is important to detect a fall incident as early as possible, whereby a slight delay in providing immediate assistance can cause severe unrecoverable injuries. One of the main challenges in fall event recognition is the imbalanced training data between fall and no-fall events, where a real-life fall incident is a sporadic event that occurs infrequently. Most of the recent techniques produce a lot of false alarms, as it is hard to train them to cover a wide range of fall situations. Hence, this paper aims to detect the exact fall frame in a video sequence, as such it will not be dependent on the whole clip of the video sequence. Our proposed approach consists of a two-stage module where the first stage employs a compact convolutional neural network tracker to generate the object trajectory information. Features of interest will be sampled from the generated trajectory paths, which will be fed as the input to the second stage. The next stage network then models the temporal dependencies of the trajectory information using symmetrical Long Short-Term Memory (LSTM) architecture. This two-stage module is a novel approach as most of the techniques rely on the detection module rather than the tracking module. The simulation experiments were tested using Fall Detection Dataset (FDD). The proposed approach obtains an expected average overlap of 0.167, which is the best performance compared to Multi-Domain Network (MDNET) and Tree-structured Convolutional Neural Network (TCNN) trackers. Furthermore, the proposed 3-layers of stacked LSTM architecture also performs the best compared to the vanilla recurrent neural network and single-layer LSTM. This approach can be further improved if the tracker model is firstly pre-tuned in offline mode with respect to a specific type of object of interest, rather than a general object.

Crowd counting from low-resolution images is a challenging task, in particular in the edge computing system. An embedded equipment is commonly incompetent at patch-based crowd counting with real-time performance. This work develops a real-time method to count passengers in a bus by using Nvidia TX2. The videos of entry are recorded by a camera up ahead, and the data suffer from severe occlusion, which makes designing handcrafted features difficult. The counting is performed by summing up pixel values of the density map estimated using a compact convolutional neural network (CCNN), which is robust to scale variations by employing skip connections. A weighted Euclidean loss is proposed to handle cluttered backgrounds and blurry foregrounds. The loss increases the activations in dense regions, but can restrain the activations in background regions. The counting results are further improved by smoothing, which utilizes constraints between consecutive frames. Comparisons with existing counting approaches, including patch-based and whole image-based approaches, are made on two benchmarking datasets. The results indicate the accuracy of CCNN in counting dense crowds. Moreover, the evaluated bus datasets verify the feasibility of CCNN in counting passengers from low-resolution input images with real-time performance on TX2.

Deep learning-based aircraft detection methods have been increasingly implemented in recent years. However, due to the multi-resolution imaging modes, aircrafts in different images show very wide diversity on size, view and other visual features, which brings great challenges to detection. Although standard deep convolution neural networks (DCNN) can extract rich semantic features, they destroy the bottom-level location information. The features of small targets may also be submerged by redundant top-level features, resulting in poor detection. To address these problems, we proposed a compact multi-scale dense convolutional neural network (MS-DenseNet) for aircraft detection in remote sensing images. Herein, DenseNet was utilized for feature extraction, which enhances the propagation and reuse of the bottom-level high-resolution features. Subsequently, we combined feature pyramid network (FPN) with DenseNet to form a MS-DenseNet for learning multi-scale features, especially features of small objects. Finally, by compressing some of the unnecessary convolution layers of each dense block, we designed three new compact architectures: MS-DenseNet-41, MS-DenseNet-65, and MS-DenseNet-77. Comparative experiments showed that the compact MS-DenseNet-65 obtained a noticeable improvement in detecting small aircrafts and achieved state-of-the-art performance with a recall of 94% and an F1-score of 92.7% and cost less computational time. Furthermore, the experimental results on robustness of UCAS-AOD and RSOD datasets also indicate the good transferability of our method.

In this paper, we carry out an assessment of cosmic distance duality relation (CDDR) based on the latest observations of HII galaxies acting as standard candles and ultra-compact structure in radio quasars acting as standard rulers. Particularly, two machine learning reconstruction methods [Gaussian Process (GP) and Artificial Neural Network (ANN)] are applied to reconstruct the Hubble diagrams from observational data. We show that both approaches are capable of reconstructing the current constraints on possible deviations from the CDDR in the redshift range z∼2.3. Considering four different parametric methods of CDDR, which quantify deviations from the CDDR and the standard cosmological model, we compare the results of the two different machine learning approaches. It is observed that the validity of CDDR is in well agreement with the current observational data within 1σ based on the reconstructed distances through GP in the overlapping redshift domain. Moreover, we find that ultra-compact radio quasars could provide 10-3-level constraints on the violation parameter at high redshifts, when combined with the observations of HII galaxies. In the framework of ANN, one could derive robust constraints on the violation parameter at a precision of 10-2, with the validity of such distance duality relation within 2σ confidence level.

Deep convolutional neural networks (DCNN) have shown promising performance in plant disease recognition. However, these networks cannot be deployed on resource-limited smart devices due to their vast parameters and computations. To address the issue of deployability when developing cotton disease identification applications for mobile/smart devices, we compress the disease recognition models employing the pruning algorithm. The algorithm uses the γ coefficient in the Batch Normalization layer to prune the channels to realize the compression of DCNN. To further improve the accuracy of the model, we suggest two strategies in combination with transfer learning: compression after transfer learning or transfer learning after compression. In our experiments, the source dataset is famous PlantVillage while the target dataset is the cotton disease image set which contains images collected from the Internet and taken from the fields. We select VGG16, ResNet164 and DenseNet40 as compressed models for comparison. The experimental results show that transfer learning after compression overall surpass its counterpart. When compression rate is set to 80% the accuracies of compressed version of VGG16, ResNet164 and DenseNet40 are 90.77%, 96.31% and 97.23%, respectively, and the parameters are only 0.30M, 0.43M and 0.26M, respectively. Among the compressed models, DenseNet40 has the highest accuracy and the smallest parameters. The best model (DenseNet40-80%-T) is pruned 75.70% of the parameters and cut off 65.52% of the computations, with the model size being only 2.2 MB. Compared with the version of compression after transfer learning, the accuracy of the model is improved by 0.74%. We further develop a cotton disease recognition APP on the Android platform based on the model and on the test phone, the average time to identify a single image is just 87ms.

Convolutional Neural Networks (CNNs) have achieved remarkable results in many application fields. However, these CNNs have a large number of network parameters, thereby consuming a lot of computation and storage resources. This makes CNNs unable to be effectively applied to these platforms with limited storage and computation resources. To address this issue, this paper proposes a new compact convolution module called DSC-Ghost-Conv, which combines the advantages of both depthwise separable convolution (DSC) and Ghost convolution module (Ghost-Conv). DSC-Ghost-Conv replaces the standard convolution used in the Ghost convolution module with depthwise separable convolution so as to reduce resource costs of the Ghost convolution module. DSC-Ghost-Conv can be used as a plug-and-play component to implement ordinary convolutional layers in typical CNNs such as VGG-16, ResNet-50 and GoogleNet. Experimental results on the MNIST and CIFAR-10 datasets show that implementing the ordinary convolutional layers of CNNs with DSC-Ghost-Conv not only obtains the competitive performance to typical CNNs, but also greatly reduces the number of network parameters and floating point operations (FLOPs) of CNNs. This demonstrates that the proposed DSC-Ghost-Conv can effectively reduce the resource costs of CNNs.

A methodology to develop artificial neural network (ANN) models to quickly incorporate the characteristics of emerging devices for circuit simulation is described in this work. To improve the model accuracy, a current and voltage data preprocessing scheme is proposed to derive a minimum dataset to train the ANN model with sufficient accuracy. To select a proper network size, four guidelines are developed from the principles of two-layer network. With that, a reference ANN size is proposed as a generic three-terminal transistor model. The ANN model formulated using the proposed approach has been verified by physical device data. Both the device and circuit-level tests show that the ANN model can reproduce and predict various device and circuits with high accuracy.

In this paper, we present an artificial neural network (ANN)-based compact model to evaluate the characteristics of a nanosheet field-effect transistor (NSFET), which has been highlighted as a next-generation nano-device. To extract data reflecting the accurate physical characteristics of NSFETs, the Sentaurus TCAD (technology computer-aided design) simulator was used. The proposed ANN model accurately and efficiently predicts currents and capacitances of devices using the five proposed key geometric parameters and two voltage biases. A variety of experiments were carried out in order to create a powerful ANN-based compact model using a large amount of data up to the sub-3-nm node. In addition, the activation function, physics-augmented loss function, ANN structure, and preprocessing methods were used for effective and efficient ANN learning. The proposed model was implemented in Verilog-A. Both a global device model and a single-device model were developed, and their accuracy and speed were compared to those of the existing compact model. The proposed ANN-based compact model simulates device characteristics and circuit performances with high accuracy and speed. This is the first time that a machine learning (ML)-based compact model has been demonstrated to be several times faster than the existing compact model.