Lightweight Convolutional Neural Network Research Articles

LiteCCLKNet: A lightweight criss‐cross large kernel convolutional neural network for hyperspectral image classification

AbstractHigh‐performance convolutional neural networks (CNNs) stack many convolutional layers to obtain powerful feature extraction capability, which leads to huge storing and computational costs. The authors focus on lightweight models for hyperspectral image (HSI) classification, so a novel lightweight criss‐cross large kernel convolutional neural network (LiteCCLKNet) is proposed. Specifically, a lightweight module containing two 1D convolutions with self‐attention mechanisms in orthogonal directions is presented. By setting large kernels within the 1D convolutional layers, the proposed module can efficiently aggregate long‐range contextual features. In addition, the authors effectively obtain a global receptive field by stacking only two of the proposed modules. Compared with traditional lightweight CNNs, LiteCCLKNet reduces the number of parameters for easy deployment to resource‐limited platforms. Experimental results on three HSI datasets demonstrate that the proposed LiteCCLKNet outperforms the previous lightweight CNNs and has higher storage efficiency.

Multi-Task Learning U-Net for Functional Shoulder Sub-Task Segmentation.

The assessment of a frozen shoulder (FS) is critical for evaluating outcomes and medical treatment. Analysis of functional shoulder sub-tasks provides more crucial information, but current manual labeling methods are time-consuming and prone to errors. To address this challenge, we propose a deep multi-task learning (MTL) U-Net to provide an automatic and reliable functional shoulder sub-task segmentation (STS) tool for clinical evaluation in FS. The proposed approach contains the main task of STS and the auxiliary task of transition point detection (TPD). For the main STS task, a U-Net architecture including an encoder-decoder with skip connection is presented to perform shoulder sub-task classification for each time point. The auxiliary TPD task uses lightweight convolutional neural networks architecture to detect the boundary between shoulder sub-tasks. A shared structure is implemented between two tasks and their objective functions of them are optimized jointly. The fine-grained transition-related information from the auxiliary TPD task is expected to help the main STS task better detect boundaries between functional shoulder sub-tasks. We conduct the experiments using wearable inertial measurement units to record 815 shoulder task sequences collected from 20 healthy subjects and 43 patients with FS. The experimental results present that the deep MTL U-Net can achieve superior performance compared to using single-task models. It shows the effectiveness of the proposed method for functional shoulder STS. The code has been made publicly available at https://github.com/RobinChu9890/MTL-U-Net-for-Functional-Shoulder-STS.Clinical Relevance- This work provides an automatic and reliable functional shoulder sub-task segmentation tool for clinical evaluation in frozen shoulder.

Machine tool fault classification diagnosis based on audio parameters

With the increasing precision, complexity, and diversification of CNC machine tools, the requirements for fault detection technology of CNC machine tools are getting essential.If problems such as tool wear and spindle locking can not be found in time, it will lead to the waste of a large number of raw materials and production time. Aiming at the fault problem of high-precision NC machine tools, this paper uses the method based on DWT and statistical domain to extract features from audio signals to form data sets, and proposes a lightweight convolution neural network to identify the fault types of machine tools.Firstly, features are extracted from MIMII public data sets, and evaluates the data sets by using three machine learning models (SVM, KNN, and GNB).The results show that when facing different materials and structures, the model trained by extracting features from mechanical audio signals based on wavelet transform and statistical domain has good prediction accuracy.Then, this method is used for audio data of CNC machine tools in a real factory environment.The data set includes six common machine tool faults, including normal operation, tool wear, spindle locking, tool idling, abnormal impact, and back cutting amount.A lightweight convolution neural network was used to identify fault types in machine tools.The experimental results show that the recognition accuracy of the trained model is 97.85%, AUC is 0.9984, and F1 is 0.9769. Compared with the DNN proposed by Jong-Yi (Kuo et al., 2022) [1] , it has been improved in all aspects. The accuracy of fault recognition is improved from 96.1% to 97.85% compared with the method of sound fusion feature and OCSVM (Ding et al., 2022) [2].

Audio features based ADS-CNN method for flight attitude recognition of quadrotor UAV

With the rapid development of unmanned flight technology, it has been widely used in all walks of life. Quad-rotor unmanned aerial vehicle (QrUAV) is the most widely used, due to its advantages of simple structure, high stability, and easy control. Currently, efficient, fast and accurate identification of QrUAV is still a hot issue. To this end, this paper proposes an audio features based flight attitude recognition method for QrUAV. Using the newly recorded UAV Attitude Audio data set (UAVAA), a lightweight convolutional neural network (ADS-CNN) integrated with attention mechanism is established for the flight attitude recognition of QrUAV. The feature of this method is that a lightweight network structure is designed through depthwise separable convolution and residual connection, this structure effectively reduces the network parameter consumption, and increases the network depth as much as possible when the number of UAVAA samples is small, which effectively suppresses overfitting and improves the recognition accuracy. In addition, by introducing of the region of interest focusing module (IFA) into the network, the weight of the attitude features is divided to realize the importance of the features, so as to achieve efficient and high-accuracy recognition. Finally, by fusing the MFCC features and STFT features of the QrUAV audio, the feature dimension is increased and the recognition accuracy is improved. By experimental comparison and analysis, the accuracy rate of ADS-CNN reached 98.81%, which was 2.7% higher than that of VGG16, the model parameters of ADS-CNN were only 27.6% of VGG16, and the operation time was only 32% of VGG16, which shows that ADS –CNN can identify the flight attitude of the QrUAV more accurately and quickly by using attitude audio.

Classification of multiple power quality disturbances based on continuous wavelet transform and lightweight convolutional neural network

AbstractAiming at the problems of noise interference and too many network parameters for power quality disturbances' (PQDs') classification based on deep learning, the lightweight convolutional neural network combining maximum likelihood Kalman filter and continuous wavelet transform is proposed. In this proposed method, the disturbed PQD signals are denoised by maximum likelihood Kalman filter, and then the denoised PQDs are converted to time‐frequency diagrams, which can provide rich time and frequency domain information, and finally the lightweight convolution neural network is used for automatically extracting and classifying multiple PQDs. To verify the effectiveness and superiority of the proposed method, a variety of PQDs were tested under different noise levels, the experiment results indicate that the average classification accuracy can reach more than 99% even in the case of 10 dB noise. Compared with the existing classification methods, the accuracy and noise immunity ability are improved. Additionally, the proposed method has decided advantages, as evidenced by its low parameter count of 0.73M and short average test time with only 0.7 ms.

Toward metacognition: subject-aware contrastive deep fusion representation learning for EEG analysis.

We propose a subject-aware contrastive learning deep fusion neural network framework for effectively classifying subjects' confidence levels in the perception of visual stimuli. The framework, called WaveFusion, is composed of lightweight convolutional neural networks for per-lead time-frequency analysis and an attention network for integrating the lightweight modalities for final prediction. To facilitate the training of WaveFusion, we incorporate a subject-aware contrastive learning approach by taking advantage of the heterogeneity within a multi-subject electroencephalogram dataset to boost representation learning and classification accuracy. The WaveFusion framework demonstrates high accuracy in classifying confidence levels by achieving a classification accuracy of 95.7% while also identifying influential brain regions.

Develop a Lightweight Convolutional Neural Network to Recognize Palms Using 3D Point Clouds

Biometrics has become an important research issue in recent years, and the use of deep learning neural networks has made it possible to develop more reliable and efficient recognition systems. Palms have been identified as one of the most promising candidates among various biometrics due to their unique features and easy accessibility. However, traditional palm recognition methods involve 3D point clouds, which can be complex and difficult to work with. To mitigate this challenge, this paper proposes two methods which are Multi-View Projection (MVP) and Light Inverted Residual Block (LIRB).The MVP simulates different angles that observers use to observe palms in reality. It transforms 3D point clouds into multiple 2D images and effectively reduces the loss of mapping 3D data to 2D data. Therefore, the MVP can greatly reduce the complexity of the system. In experiments, MVP demonstrated remarkable performance on various famous models, such as VGG or MobileNetv2, with a particular improvement in the performance of smaller models. To further improve the performance of small models, this paper applies LIRB to build a lightweight 2D CNN called Tiny-MobileNet (TMBNet).The TMBNet has only a few convolutional layers but outperforms the 3D baselines PointNet and PointNet++ in FLOPs and accuracy. The experimental results show that the proposed method can effectively mitigate the challenges of recognizing palms through 3D point clouds of palms. The proposed method not only reduces the complexity of the system but also extends the use of lightweight CNN. These findings have significant implications for developing biometrics and could lead to improvements in various fields, such as access control and security control.

A Ship Tracking and Speed Extraction Framework in Hazy Weather Based on Deep Learning

Obtaining ship navigation information from maritime videos can significantly improve maritime supervision efficiency and enable timely safety warnings. Ship detection and tracking are essential technologies for mining video information. However, current research focused on these advanced vision tasks in maritime supervision is not sufficiently comprehensive. Taking into account the application of ship detection and tracking technology, this study proposes a deep learning-based ship speed extraction framework under the haze environment. First, a lightweight convolutional neural network (CNN) is used to remove haze from images. Second, the YOLOv5 algorithm is used to detect ships in dehazed marine images, and a simple online and real-time tracking method with a Deep association metric (Deep SORT) is used to track ships. Then, the ship’s displacement in the images is calculated based on the ship’s trajectory. Finally, the speed of the ships is estimated by calculating the mapping relationship between the image space and real space. Experiments demonstrate that the method proposed in this study effectively reduces haze interference in maritime videos, thereby enhancing the image quality while extracting the ship’s speed. The mean squared error (MSE) for multiple scenes is 0.3 Kn on average. The stable extraction of ship speed from the video achieved in this study holds significant value in further ensuring the safety of ship navigation.

Handwritten Bengali Alphabets, Compound Characters and Numerals Recognition Using CNN-based Approach

Accurately classifying user-independent handwritten Bengali characters and numerals presents a formidable challenge in their recognition. This task becomes more complicated due to the inclusion of numerous complex-shaped compound characters and the fact that different authors employ diverse writing styles. Researchers have recently conducted significant researches using individual approaches to recognize handwritten Bangla digits, alphabets, and slightly compound characters. To address this, we propose a straightforward and lightweight convolutional neural network (CNN) framework to accurately categorize handwritten Bangla simple characters, compound characters, and numerals. The suggested approach exhibits outperformance in terms of performance when compared too many previously developed procedures, with faster execution times and requiring fewer epochs. Furthermore, this model applies to more than three datasets. Our proposed CNN-based model has achieved impressive validation accuracies on three datasets. Specifically, for the BanglaLekha isolated dataset, which includes 84-character classes, the validation accuracy was 92.48%. On the Ekush dataset, which includes 60-character classes, the model achieved a validation accuracy of 97.24%, while on the customized dataset, which includes 50-character classes, the validation accuracy was 97.03%. Our model has demonstrated high accuracy and outperformed several prominent existing frameworks.

Estimation of number of unmanned aerial vehicles in a scene utilizing acoustic signatures and machine learning.

With the exponential growth in unmanned aerial vehicle (UAV)-based applications, there is a need to ensure safe and secure operations. From a security perspective, detecting and localizing intruder UAVs is still a challenge. It is even more challenging to accurately estimate the number of intruder UAVs on the scene. In this work, we propose a simple acoustic-based technique to detect and estimate the number of UAVs. Our method utilizes acoustic signals generated from the motion of UAV motors and propellers. Acoustic signals are captured by flying an arbitrary number of ten UAVs in different combinations in an indoor setting. The recorded acoustic signals are trimmed, processed, and arranged to create an UAV audio dataset. The UAV audio dataset is subjected to time-frequency transformations to generate audio spectrogram images. The generated spectrogram images are then fed to a custom lightweight convolutional neural network (CNN) architecture to estimate the number of UAVs in the scene. Following training, the proposed model achieves an average test accuracy of 93.33% as compared to state-of-the-art benchmark models. Furthermore, the deployment feasibility of the proposed model is validated by running inference time calculations on edge computing devices, such as the Raspberry Pi 4, NVIDIA Jetson Nano, and NVIDIA Jetson AGX Xavier.

A Leading but Simple Classification Method for Remote Sensing Images

Recently, researchers have proposed a lot of deep convolutional neural network (CNN) approaches with obvious flaws to tackle the difficult semantic classification (SC) task of remote sensing images (RSI). In this paper, the author proposes a simple method that aims to provide a leading but efficient solution by using a lightweight EfficientNet-B0. First, this paper concluded the drawbacks with an analysis of mathematical theory and then proposed a qualitative conclusion on the previous methods’ theoretical performance based on theoretical derivation and experiments. Following that, the paper designs a novel method named LS-EfficientNet, consisting only of a single CNN and a concise training algorithm called SC-CNN. Far different from previous complex and hardware-extensive ones, the proposed method mainly focuses on tackling the long-neglected problems, including overfitting, data distribution shift by DA, improper use of training tricks, and other incorrect operations on a pre-trained CNN. Compared to previous studies, the proposed method is easy to reproduce because all the models, training tricks, and hyperparameter settings are open-sourced. Extensive experiments on two benchmark datasets show that the proposed method can easily surpass all the previous state-of-the-art ones, with an outstanding accuracy lead of 0.5% to 1.2% and a remarkable parameter decrease of 78% if compared to the best prior one in 2022. In addition, ablation test results also prove that the proposed effective combination of training tricks, including OLS and CutMix, can clearly boost a CNN's performance for RSI-SC, with an increase in accuracy of 1.0%. All the results reveal that a single lightweight CNN can well tackle the routine task of classifying RSI.

A Lightweight Convolutional Neural Network Based on Dynamic Level-Set Loss Function for Spine MR Image Segmentation.

Spine MR image segmentation is important foundation for computer-aided diagnostic (CAD) algorithms of spine disorders. Convolutional neural networks segment effectively, but require high computational costs. To design a lightweight model based on dynamic level-set loss function for high segmentation performance. Retrospective. Four hundred forty-eight subjects (3163 images) from two separate datasets. Dataset-1: 276 subjects/994 images (53.26% female, mean age 49.02 ± 14.09), all for disc degeneration screening, 188 had disc degeneration, 67 had herniated disc. Dataset-2: public dataset with 172 subjects/2169 images, 142 patients with vertebral degeneration, 163 patients with disc degeneration. T2 weighted turbo spin echo sequences at 3T. Dynamic Level-set Net (DLS-Net) was compared with four mainstream (including U-net++) and four lightweight models, and manual label made by five radiologists (vertebrae, discs, spinal fluid) used as segmentation evaluation standard. Five-fold cross-validation are used for all experiments. Based on segmentation, a CAD algorithm of lumbar disc was designed for assessing DLS-Net's practicality, and the text annotation (normal, bulging, or herniated) from medical history data were used as evaluation standard. All segmentation models were evaluated with DSC, accuracy, precision, and AUC. The pixel numbers of segmented results were compared with manual label using paired t-tests, with P < 0.05 indicating significance. The CAD algorithm was evaluated with accuracy of lumbar disc diagnosis. With only 1.48% parameters of U-net++, DLS-Net achieved similar accuracy in both datasets (Dataset-1: DSC 0.88 vs. 0.89, AUC 0.94 vs. 0.94; Dataset-2: DSC 0.86 vs. 0.86, AUC 0.93 vs. 0.93). The segmentation results of DLS-Net showed no significant differences with manual labels in pixel numbers for discs (Dataset-1: 1603.30 vs. 1588.77, P = 0.22; Dataset-2: 863.61 vs. 886.4, P = 0.14) and vertebrae (Dataset-1: 3984.28 vs. 3961.94, P = 0.38; Dataset-2: 4806.91 vs. 4732.85, P = 0.21). Based on DLS-Net's segmentation results, the CAD algorithm achieved higher accuracy than using non-cropped MR images (87.47% vs. 61.82%). The proposed DLS-Net has fewer parameters but achieves similar accuracy to U-net++, helps CAD algorithm achieve higher accuracy, which facilitates wider application. 2 TECHNICAL EFFICACY: Stage 1.

Individualized Indicators and Estimation Methods for Tiger Nut (Cyperus esculentus L.) Tubers Yield Using Light Multispectral UAV and Lightweight CNN Structure

Tiger nuts are a non-genetically modified organism crop with high adaptability and economic value, and they are being widely promoted for cultivation in China. This study proposed a new yield-estimation method based on a lightweight convolutional neural network (CNN) named Squeeze Net to provide accurate production forecasts for tiger nut tubers. The multispectral unmanned aerial vehicle (UAV) images were used to establish phenotypic datasets of tiger nuts, comprising vegetation indices (VIs) and plant phenotypic indices. The Squeeze Net model with a lightweight CNN structure was constructed to fully explore the explanatory power of the spectral UAV-derived information and compare the differences between the parametric and nonparametric models applied in tiger nut yield predictions. Compared with stepwise multiple linear regression (SMLR), both algorithms achieved good yield prediction performances. The highest obtained accuracies reflected an R2 value of 0.775 and a root-mean-square error (RMSE) value of 688.356 kg/ha with SMLR, and R2 = 0.780 and RMSE = 716.625 kg/ha with Squeeze Net. This study demonstrated that Squeeze Net can efficiently process UAV multispectral images and improve the resolution and accuracy of the yield prediction results. Our study demonstrated the enormous potential of artificial intelligence (AI) algorithms in the precise crop management of tiger nuts in the arid sandy lands of northwest China by exploring the interactions between various intensive phenotypic traits and productivity.

An embarrassingly simple approach for visual navigation of forest environments.

Navigation in forest environments is a challenging and open problem in the area of field robotics. Rovers in forest environments are required to infer the traversability of a priori unknown terrains, comprising a number of different types of compliant and rigid obstacles, under varying lighting and weather conditions. The challenges are further compounded for inexpensive small-sized (portable) rovers. While such rovers may be useful for collaboratively monitoring large tracts of forests as a swarm, with low environmental impact, their small-size affords them only a low viewpoint of their proximal terrain. Moreover, their limited view may frequently be partially occluded by compliant obstacles in close proximity such as shrubs and tall grass. Perhaps, consequently, most studies on off-road navigation typically use large-sized rovers equipped with expensive exteroceptive navigation sensors. We design a low-cost navigation system tailored for small-sized forest rovers. For navigation, a light-weight convolution neural network is used to predict depth images from RGB input images from a low-viewpoint monocular camera. Subsequently, a simple coarse-grained navigation algorithm aggregates the predicted depth information to steer our mobile platform towards open traversable areas in the forest while avoiding obstacles. In this study, the steering commands output from our navigation algorithm direct an operator pushing the mobile platform. Our navigation algorithm has been extensively tested in high-fidelity forest simulations and in field trials. Using no more than a 16 × 16 pixel depth prediction image from a 32 × 32 pixel RGB image, our algorithm running on a Raspberry Pi was able to successfully navigate a total of over 750 m of real-world forest terrain comprising shrubs, dense bushes, tall grass, fallen branches, fallen tree trunks, small ditches and mounds, and standing trees, under five different weather conditions and four different times of day. Furthermore, our algorithm exhibits robustness to changes in the mobile platform's camera pitch angle, motion blur, low lighting at dusk, and high-contrast lighting conditions.

A multi-feature fusion algorithm for driver fatigue detection based on a lightweight convolutional neural network

A multi-feature fusion algorithm for driver fatigue detection based on a lightweight convolutional neural network

A High Performance Reconfigurable Hardware Architecture for Lightweight Convolutional Neural Network

Since the lightweight convolutional neural network EfficientNet was proposed by Google in 2019, the series of models have quickly become very popular due to their superior performance with a small number of parameters. However, the existing convolutional neural network hardware accelerators for EfficientNet still have much room to improve the performance of the depthwise convolution, squeeze-and-excitation module and nonlinear activation functions. In this paper, we first design a reconfigurable register array and computational kernel to accelerate the depthwise convolution. Next, we propose a vector unit to implement the nonlinear activation functions and the scale operation. An exchangeable-sequence dual-computational kernel architecture is proposed to improve the performance and the utilization. In addition, the memory architectures are designed to complete the hardware accelerator for the above computing architecture. Finally, in order to evaluate the performance of the hardware accelerator, the accelerator is implemented based on Xilinx XCVU37P. The results show that the proposed accelerator can work at the main system clock frequency of 300 MHz with the DSP kernel at 600 MHz. The performance of EfficientNet-B3 in our architecture can reach 69.50 FPS and 255.22 GOPS. Compared with the latest EfficientNet-B3 accelerator, which uses the same FPGA development board, the accelerator proposed in this paper can achieve a 1.28-fold improvement of single-core performance and 1.38-fold improvement of performance of each DSP.

Detection of various lung diseases including COVID-19 using extreme learning machine algorithm based on the features extracted from a lightweight CNN architecture

Around the world, several lung diseases such as pneumonia, cardiomegaly, and tuberculosis (TB) contribute to severe illness, hospitalization or even death, particularly for elderly and medically vulnerable patients. In the last few decades, several new types of lung-related diseases have taken the lives of millions of people, and COVID-19 has taken almost 6.27 million lives. To fight against lung diseases, timely and correct diagnosis with appropriate treatment is crucial in the current COVID-19 pandemic. In this study, an intelligent recognition system for seven lung diseases has been proposed based on machine learning (ML) techniques to aid the medical experts. Chest X-ray (CXR) images of lung diseases were collected from several publicly available databases. A lightweight convolutional neural network (CNN) has been used to extract characteristic features from the raw pixel values of the CXR images. The best feature subset has been identified using the Pearson Correlation Coefficient (PCC). Finally, the extreme learning machine (ELM) has been used to perform the classification task to assist faster learning and reduced computational complexity. The proposed CNN-PCC-ELM model achieved an accuracy of 96.22% with an Area Under Curve (AUC) of 99.48% for eight class classification. The outcomes from the proposed model demonstrated better performance than the existing state-of-the-art (SOTA) models in the case of COVID-19, pneumonia, and tuberculosis detection in both binary and multiclass classifications. For eight class classification, the proposed model achieved precision, recall and fi-score and ROC are 100%, 99%, 100% and 99.99% respectively for COVID-19 detection demonstrating its robustness. Therefore, the proposed model has overshadowed the existing pioneering models to accurately differentiate COVID-19 from the other lung diseases that can assist the medical physicians in treating the patient effectively.

MGIC: Multigrid-in-Channels Neural Network Architectures

We present a multigrid-in-channels (MGIC) approach that tackles the quadratic growth of the number of parameters with respect to the number of channels in standard convolutional neural networks (CNNs). Thereby our approach addresses the redundancy in CNNs that is also exposed by the recent success of lightweight CNNs. Lightweight CNNs can achieve comparable accuracy to standard CNNs with fewer parameters; however, the number of weights still scales quadratically with the CNN’s width. Our MGIC architectures replace each CNN block with an MGIC counterpart that utilizes a hierarchy of nested grouped convolutions of small group size to address this. Hence, our proposed architectures scale linearly with respect to the network’s width while retaining full coupling of the channels as in standard CNNs. Our extensive experiments on image classification, segmentation, and point cloud classification show that applying this strategy to different architectures like ResNet and MobileNetV3 reduces the number of parameters while obtaining similar or better accuracy.

High-Dynamic-Range Tone Mapping in Intelligent Automotive Systems

Intelligent driver assistance systems are becoming increasingly popular in modern passenger vehicles. A crucial component of intelligent vehicles is the ability to detect vulnerable road users (VRUs) for an early and safe response. However, standard imaging sensors perform poorly in conditions of strong illumination contrast, such as approaching a tunnel or at night, due to their dynamic range limitations. In this paper, we focus on the use of high-dynamic-range (HDR) imaging sensors in vehicle perception systems and the subsequent need for tone mapping of the acquired data into a standard 8-bit representation. To our knowledge, no previous studies have evaluated the impact of tone mapping on object detection performance. We investigate the potential for optimizing HDR tone mapping to achieve a natural image appearance while facilitating object detection of state-of-the-art detectors designed for standard dynamic range (SDR) images. Our proposed approach relies on a lightweight convolutional neural network (CNN) that tone maps HDR video frames into a standard 8-bit representation. We introduce a novel training approach called detection-informed tone mapping (DI-TM) and evaluate its performance with respect to its effectiveness and robustness in various scene conditions, as well as its performance relative to an existing state-of-the-art tone mapping method. The results show that the proposed DI-TM method achieves the best results in terms of detection performance metrics in challenging dynamic range conditions, while both methods perform well in typical, non-challenging conditions. In challenging conditions, our method improves the detection F2 score by 13%. Compared to SDR images, the increase in F2 score is 49%.

Research on image classification of coal and gangue based on a lightweight convolution neural network

AbstractIt is vital to differentiate between coal and gangue in coal mining effectively. In recent years, classification methods for images of coal‐ and gangue‐based convolutional neural networks have emerged, which makes the analysis process simple and fast. However, these methods still have the limitation of balancing recognition performance and computational efficiency due to the limitation of device hardware. This research proposes a group convolution and channel shuffle augmentation Ghost network (GSAGNet) for the classification task of coal and gangue images. First, we design a new Ghost module with group convolution and channel shuffle to improve the stability of the Ghost module for better extraction of features of coal and gangue. Second, we redesign GhostNet to improve the model's classification accuracy while guaranteeing similar floating‐point operations. Finally, an efficient channel attention module is embedded in the prediction output of the model to improve the model's prediction results further. Extensive experiments on actual image datasets of coal and gangue show the feasibility and superiority of our proposed GSAG module. The GSAG module is applied to models such as AlexNet, VGG13, and ResNet50, and the accuracy of the models is improved by 0.27%, 1.24%, and 1.25%, respectively, compared to the Ghost module. The classification accuracy and F1‐score of the proposed GSAGNet model reach 97.50% and 97.50%, respectively, which are 1.01% and 1.04% higher than that of GhostNet. The GSAGNet model in this study can classify coal and gangue efficiently, which can effectively enable the separation of underground coal and gangue and has specific practical application significance and value.