Popular Convolutional Neural Networks Research Articles

UNIFIED MULTIMODAL BIOMETRICS FUSION USING DEEP LEARNING FOR SECURING IOT

Advancements in multimodal biometrics, which amalgamate multiple biometric traits, hold promise for augmenting the accuracy and robustness of biometric identification systems. The focal point of this innovative study is the enhancement of multimodal biometrics identification, using face and iris images as the key biometric traits. This work taps into the expansive collection of face and iris images present in the WVU-Multimodal dataset for evaluation purposes. Our proposed approach employs “Convolutional Neural Network (CNN)” architectures, notable for their efficacy in computer vision tasks, to extract potent discriminative features from the input images. This work specifically incorporates three popular CNN architectures: ResNet-50, InceptionNet, XceptionNet, and fine-tuned CNN. To amalgamate the extracted features, investigate various fusion techniques in the security-centric industry: early fusion, and score-level fusion. Early fusion is an approach that merges the raw images of both face and iris at the input level to a single CNN model. Use the Gabor approach to enhance the image's quality and make the face and iris information more visible. This technique modifies the histogram equalization process for local regions, thus enabling better visibility and subsequent feature extraction. Our experimental evaluation employs performance metrics like accuracy, “Equal Error Rate”, and “Receiver Operating Characteristic” curves. In this work undertakes a comparative analysis to appraise the performance of the different CNN architectures and fusion techniques under scrutiny.

CNN Pruning with Multi-Stage Feature Decorrelation

This paper proposes a channel pruning method based on multi-stage feature de-correlation to obtain a more efficient convolutional neural network (CNN) model. Based on the correlation of hidden features at each level of the network, we refine more efficient features of each convolutional layer by applying feature de-correlation constraints (MFD Loss) to each convolutional layer of the network and then prune channels according to the modulus of the feature maps output from each layer. After several rounds of pruning and fine-tuning, a network with similar accuracy, a substantially smaller network size, and more efficient operation is generated compared to the original model. Our experiments on pruning various popular CNN models on many standard datasets demonstrate the method’s effectiveness. Specifically, for VGG-16 on CIFAR10, our approach eliminates parameters by 97.0%, saves Float-Point-Operations (FLOPs) by 66.9%, with a 0.4% accuracy gain and state-of-art performance. For ResNet-50 on the ImageNet dataset, our method eliminates parameters by 30.0%, and saves FLOPs by 52%, with 1.4% accuracy loss, which also proves the effectiveness of the method. The code for the paper can be found at https://github.com/lovelyemperor/MFD .

DeepEMPR: coffee leaf disease detection with deep learning and enhanced multivariance product representation.

Plant diseases threaten agricultural sustainability by reducing crop yields. Rapid and accurate disease identification is crucial for effective management. Recent advancements in artificial intelligence (AI) have facilitated the development of automated systems for disease detection. This study focuses on enhancing the classification of diseases and estimating their severity in coffee leaf images. To do so, we propose a novel approach as the preprocessing step for the classification in which enhanced multivariance product representation (EMPR) is used to decompose the considered image into components, a new image is constructed using some of those components, and the contrast of the new image is enhanced by applying high-dimensional model representation (HDMR) to highlight the diseased parts of the leaves. Popular convolutional neural network (CNN) architectures, including AlexNet, VGG16, and ResNet50, are evaluated. Results show that VGG16 achieves the highest classification accuracy of approximately 96%, while all models perform well in predicting disease severity levels, with accuracies exceeding 85%. Notably, the ResNet50 model achieves accuracy levels surpassing 90%. This research contributes to the advancement of automated crop health management systems.

Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions

Overhead transmission line insulators are non-conductive materials that separate conductors from grounded transmission towers. Once in operation, they frequently experience environmental pollution and electrical or mechanical stress. Since adverse operational conditions can lead to insulation failure, regular inspections are essential to prevent power outages. To this end, this paper proposes a novel technique based on deep convolutional neural networks (CNNs) to classify high-voltage insulator surface conditions based on their image. Successful applications of CNNs in computer vision have led to several pretrained architectures for image classification. To use these pretrained models, a practitioner typically fine-tunes and selects one final model via a model selection stage and discards all other models. In contrast with many existing studies that use such a “winner-takes-all” approach, here, we identify the best subset of seven popular pretrained CNN architectures that are combined by soft voting to form an ensemble classifier. From a machine learning (ML) perspective, this focus is warranted because the convolutional base of each pretrained architecture operates as a feature extractor and an ensemble of them works as a combination of various feature extraction rules. Our numerical experiments demonstrate the advantage of the identified ensemble model to individual pretrained architectures.

Multimodal biometric identification: leveraging convolutional neural network (CNN) architectures and fusion techniques with fingerprint and finger vein data.

Advancements in multimodal biometrics, which integrate multiple biometric traits, promise to enhance the accuracy and robustness of identification systems. This study focuses on improving multimodal biometric identification by using fingerprint and finger vein images as the primary traits. We utilized the "NUPT-FPV" dataset, which contains a substantial number of finger vein and fingerprint images, which significantly aided our research. Convolutional neural networks (CNNs), renowned for their efficacy in computer vision tasks, are used in our model to extract distinct discriminative features. Specifically, we incorporate three popular CNN architectures: ResNet, VGGNet, and DenseNet. We explore three fusion strategies used in security applications: early fusion, late fusion, and score-level fusion. Early fusion integrates raw images at the input layer of a single CNN, combining information at the initial stages. Late fusion, in contrast, merges features after individual learning from each CNN model. Score-level fusion employs weighted aggregation to combine scores from each modality, leveraging the complementary information they provide. We also use contrast limited adaptive histogram equalization (CLAHE) to enhance fingerprint contrast and vein pattern features, improving feature visibility and extraction. Our evaluation metrics include accuracy, equal error rate (EER), and ROC curves. The fusion of CNN architectures and enhancement methods shows promising performance in identifying multimodal biometrics, aiming to increase identification accuracy. The proposed model offers a reliable authentication system using multiple biometrics to verify identity.

Multi-category sorting of plastic waste using Swin Transformer: A vision-based approach

Sorting out plastic waste (PW) from municipal solid waste (MSW) by material type is crucial for reutilization and pollution reduction. However, current automatic separation methods are costly and inefficient, necessitating an advanced sorting process to ensure high feedstock purity. This study introduces a Swin Transformer-based model for effectively detecting PW in real-world MSW streams, leveraging both morphological and material properties. And, a dataset comprising 3560 optical images and infrared spectra data was created to support this task. This vision-based system can localize and classify PW into five categories: polypropylene (PP), polyethylene (PE), polyethylene terephthalate (PET), polyvinyl chloride (PVC), and polystyrene (PS). Performance evaluations reveal an accuracy rate of 99.75% and a mean Average Precision (mAP50) exceeding 91%. Compared to popular convolutional neural network (CNN)-based models, this well-trained Swin Transformer-based model offers enhanced convenience and performance in five-category PW detection task, maintaining a mAP50 over 80% in the real-life deployment. The model's effectiveness is further supported by visualization of detection results on MSW streams and principal component analysis of classification scores. These results demonstrate the system's significant effectiveness in both lab-scale and real-life conditions, aligning with global regulations and strategies that promote innovative technologies for plastic recycling, thereby contributing to the development of a sustainable circular economy.

Fourier Convolution Block with global receptive field for MRI reconstruction

Reconstructing images from under-sampled Magnetic Resonance Imaging (MRI) signals significantly reduces scan time and improves clinical practice. However, Convolutional Neural Network (CNN)-based methods, while demonstrating great performance in MRI reconstruction, may face limitations due to their restricted receptive field (RF), hindering the capture of global features. This is particularly crucial for reconstruction, as aliasing artifacts are distributed globally. Recent advancements in Vision Transformers have further emphasized the significance of a large RF. In this study, we proposed a novel global Fourier Convolution Block (FCB) with whole image RF and low computational complexity by transforming the regular spatial domain convolutions into frequency domain. Visualizations of the effective RF and trained kernels demonstrated that FCB improves the RF of reconstruction models in practice. The proposed FCB was evaluated on four popular CNN architectures using brain and knee MRI datasets. Models with FCB achieved superior PSNR and SSIM than baseline models and exhibited more details and texture recovery. The code is publicly available at https://github.com/Haozhoong/FCB.

Bayesian optimized multimodal deep hybrid learning approach for tomato leaf disease classification

Manual identification of tomato leaf diseases is a time-consuming and laborious process that may lead to inaccurate results without professional assistance. Therefore, an automated, early, and precise leaf disease recognition system is essential for farmers to ensure the quality and quantity of tomato production by providing timely interventions to mitigate disease spread. In this study, we have proposed seven robust Bayesian optimized deep hybrid learning models leveraging the synergy between deep learning and machine learning for the automated classification of ten types of tomato leaves (nine diseased and one healthy). We customized the popular Convolutional Neural Network (CNN) algorithm for automatic feature extraction due to its ability to capture spatial hierarchies of features directly from raw data and classical machine learning techniques [Random Forest (RF), XGBoost, GaussianNB (GNB), Support Vector Machines (SVM), Multinomial Logistic Regression (MLR), K-Nearest Neighbor (KNN)], and stacking for classifications. Additionally, the study incorported a Boruta feature filtering layer to capture the statistically significant features. The standard, research-oriented PlantVillage dataset was used for the performance testing, which facilitates benchmarking against prior research and enables meaningful comparisons of classification performance across different approaches. We utilized a variety of statistical classification metrics to demonstrate the robustness of our models. Using the CNN-Stacking model, this study achieved the highest classification performance among the seven hybrid models. On an unseen dataset, this model achieved average precision, recall, f1-score, mcc, and accuracy values of 98.527%, 98.533%, 98.527%, 98.525%, and 98.268%, respectively. Our study requires only 0.174 s of testing time to correctly identify noisy, blurry, and transformed images. This indicates our approach's time efficiency and generalizability in images captured under challenging lighting conditions and with complex backgrounds. Based on the comparative analysis, our approach is superior and computationally inexpensive compared to the existing studies. This work will aid in developing a smartphone app to offer farmers a real-time disease diagnosis tool and management strategies.

Detection of malignant breast tissue using SAR observation with microwave imaging and convolutional neural network

The research paper introduces a deep learning approach for distinguishing between benign and malignant breast tissues using Ultra Wide Band (UWB) microwave imaging. A multi-resonant monopole microstrip patch antenna was crafted and placed on a female breast phantom model containing a tumor. The phantom’s dielectric properties were adjusted to mimic those of benign and malignant tissues. The Specific Absorption Rate (SAR) was analyzed by rotating the antenna at various angles throughout the phantom’s trajectory. Images capturing the SAR distribution over the breast phantom at different antenna resonances were obtained. The popular Convolutional Neural Network (CNN) based AlexNet model was employed for autonomous feature learning from SAR images. To address the challenge of overfitting with a limited dataset, the study utilized image augmentation and transfer learning methods. The proposed model achieved an impressive 98.59% accuracy in classifying benign and malignant breast phantom images, surpassing the performance of three other CNN models—VggNet16, GoogleNet, and ResNet. Notably, this microwave imaging system based on SAR images detected tumors as small as 1 mm with an accuracy of 97.08% outperforming existing malignancy screening techniques.

VerFormer: Vertebrae-Aware Transformer for Automatic Spine Segmentation from CT Images.

The accurate and efficient segmentation of the spine is important in the diagnosis and treatment of spine malfunctions and fractures. However, it is still challenging because of large inter-vertebra variations in shape and cross-image localization of the spine. In previous methods, convolutional neural networks (CNNs) have been widely applied as a vision backbone to tackle this task. However, these methods are challenged in utilizing the global contextual information across the whole image for accurate spine segmentation because of the inherent locality of the convolution operation. Compared with CNNs, the Vision Transformer (ViT) has been proposed as another vision backbone with a high capacity to capture global contextual information. However, when the ViT is employed for spine segmentation, it treats all input tokens equally, including vertebrae-related tokens and non-vertebrae-related tokens. Additionally, it lacks the capability to locate regions of interest, thus lowering the accuracy of spine segmentation. To address this limitation, we propose a novel Vertebrae-aware Vision Transformer (VerFormer) for automatic spine segmentation from CT images. Our VerFormer is designed by incorporating a novel Vertebrae-aware Global (VG) block into the ViT backbone. In the VG block, the vertebrae-related global contextual information is extracted by a Vertebrae-aware Global Query (VGQ) module. Then, this information is incorporated into query tokens to highlight vertebrae-related tokens in the multi-head self-attention module. Thus, this VG block can leverage global contextual information to effectively and efficiently locate spines across the whole input, thus improving the segmentation accuracy of VerFormer. Driven by this design, the VerFormer demonstrates a solid capacity to capture more discriminative dependencies and vertebrae-related context in automatic spine segmentation. The experimental results on two spine CT segmentation tasks demonstrate the effectiveness of our VG block and the superiority of our VerFormer in spine segmentation. Compared with other popular CNN- or ViT-based segmentation models, our VerFormer shows superior segmentation accuracy and generalization.

Deep Learning-Based Disease Detection in Sugarcane Leaves: Evaluating EfficientNet Models

Sugarcane is a crucial agricultural crop, providing 75% of the world's sugar production. Like all plant species, any disease that affects sugarcane can significantly impact yield and planning. Traditional manual methods for diagnosing diseases in sugarcane leaves are slow, inefficient, and often lack accuracy. In this study, we present a deep learning-based approach for the robust detection of diseases in sugarcane leaves. Specifically, we trained and evaluated all models from the EfficientNetv1 and EfficientNetv2 architectures, which are among the most notable convolutional neural network (CNN) architectures, using the publicly available Sugarcane Leaf Dataset. This dataset includes 11 disease classes and a total of 6748 images. Additionally, we compared these models with other popular CNN models. Our findings reveal that there is no direct correlation between model complexity, depth, and accuracy for the 11-class sugarcane dataset. Among the 13 models tested, EfficientNet-b6 and InceptionV4 achieved the highest accuracy rates of 93.39% and 93.10%, respectively. These results have a big impact on the ways managers can manage diseases and the agricultural processes of sugarcane production. A deep learning-based disease detection system facilitating the diagnostic process can, in turn, result in a more accurate and faster identification of diseases. That may enable farmers and agricultural managers to make timely and informed decisions, cutting down crop loss and enhancing the whole yield. These highlight the potential of deep learning in developing fast, accurate, and automatic disease diagnosis systems, which can significantly improve disease management and increase the yield of sugarcane.

Deep learning for mango leaf disease identification: A vision transformer perspective

Over the last decade, the use of machine learning in smart agriculture has surged in popularity. Deep learning, particularly Convolutional Neural Networks (CNNs), has been useful in identifying diseases in plants at an early stage. Recently, Vision Transformers (ViTs) have proven to be effective in image classification tasks. These architectures often outperform most state-of-the-art CNN models. However, the adoption of vision transformers in agriculture is still in its infancy. In this paper, we evaluated the performance of vision transformers in identification of mango leaf diseases and compare them with popular CNNs. We proposed an optimized model based on a pretrained Data-efficient Image Transformer (DeiT) architecture that achieves 99.75% accuracy, better than many popular CNNs including SqueezeNet, ShuffleNet, EfficientNet, DenseNet121, and MobileNet. We also demonstrated that vision transformers can have a shorter training time than CNNs, as they require fewer epochs to achieve optimal results. We also proposed a mobile app that uses the model as a backend to identify mango leaf diseases in real-time.

Revealing CNN Architectures via Side-Channel Analysis in Dataflow-based Inference Accelerators

Convolutional Neural Networks (CNNs) are widely used in various domains, including image recognition, medical diagnosis and autonomous driving. Recent advances in dataflow-based CNN accelerators have enabled CNN inference in resource-constrained edge devices. These dataflow accelerators utilize inherent data reuse of convolution layers to process CNN models efficiently. Concealing the architecture of CNN models is critical for privacy and security. This paper evaluates memory-based side-channel information to recover CNN architectures from dataflow-based CNN inference accelerators. The proposed attack exploits spatial and temporal data reuse of the dataflow mapping on CNN accelerators and architectural hints to recover the structure of CNN models. Experimental results demonstrate that our proposed side-channel attack can recover the structures of popular CNN models, namely Lenet, Alexnet, VGGnet16, and YOLOv2.

Computational experiments with cellular-automata generated images reveal intrinsic limitations of convolutional neural networks on pattern recognition tasks

The extraordinary success of convolutional neural networks (CNNs) in various computer vision tasks has revitalized the field of artificial intelligence. The out-sized expectations created by this extraordinary success have, however, been tempered by a recognition of CNNs’ fragility. Importantly, the magnitude of the problem is unclear due to a lack of rigorous benchmark datasets. Here, we propose a solution to the benchmarking problem that reveals the extent of the vulnerabilities of CNNs and of the methods used to provide interpretability to their predictions. We employ cellular automata (CA) to generate images with rigorously controllable characteristics. CA allow for the definition of both extraordinarily simple and highly complex discrete functions and allow for the generation of boundless datasets of images without repeats. In this work, we systematically investigate the fragility and interpretability of the three popular CNN architectures using CA-generated datasets. We find a sharp transition from a learnable phase to an unlearnable phase as the latent space entropy of the discrete CA functions increases. Furthermore, we demonstrate that shortcut learning is an inherent trait of CNNs. Given a dataset with an easy-to-learn and strongly predictive pattern, CNN will consistently learn the shortcut even if the pattern occurs only on a small fraction of the image. Finally, we show that widely used attribution methods aiming to add interpretability to CNN outputs are strongly CNN-architecture specific and vary widely in their ability to identify input regions of high importance to the model. Our results provide significant insight into the limitations of both CNNs and the approaches developed to add interpretability to their predictions and raise concerns about the types of tasks that should be entrusted to them.

BAOS-CNN: A novel deep neuroevolution algorithm for multispecies seagrass detection.

Deep learning, a subset of machine learning that utilizes neural networks, has seen significant advancements in recent years. These advancements have led to breakthroughs in a wide range of fields, from natural language processing to computer vision, and have the potential to revolutionize many industries or organizations. They have also demonstrated exceptional performance in the identification and mapping of seagrass images. However, these deep learning models, particularly the popular Convolutional Neural Networks (CNNs) require architectural engineering and hyperparameter tuning. This paper proposes a Deep Neuroevolutionary (DNE) model that can automate the architectural engineering and hyperparameter tuning of CNNs models by developing and using a novel metaheuristic algorithm, named 'Boosted Atomic Orbital Search (BAOS)'. The proposed BAOS is an improved version of the recently proposed Atomic Orbital Search (AOS) algorithm which is based on the principle of atomic model and quantum mechanics. The proposed algorithm leverages the power of the Lévy flight technique to boost the performance of the AOS algorithm. The proposed DNE algorithm (BAOS-CNN) is trained, evaluated and compared with six popular optimisation algorithms on a patch-based multi-species seagrass dataset. This proposed BAOS-CNN model achieves the highest overall accuracy (97.48%) among the seven evolutionary-based CNN models. The proposed model also achieves the state-of-the-art overall accuracy of 92.30% and 93.5% on the publicly available four classes and five classes version of the 'DeepSeagrass' dataset, respectively. This multi-species seagrass dataset is available at: https://ro.ecu.edu.au/datasets/141/.

Enhancing Skin Cancer Diagnosis Using Swin Transformer with Hybrid Shifted Window-Based Multi-head Self-attention and SwiGLU-Based MLP

Skin cancer is one of the most frequently occurring cancers worldwide, and early detection is crucial for effective treatment. Dermatologists often face challenges such as heavy data demands, potential human errors, and strict time limits, which can negatively affect diagnostic outcomes. Deep learning–based diagnostic systems offer quick, accurate testing and enhanced research capabilities, providing significant support to dermatologists. In this study, we enhanced the Swin Transformer architecture by implementing the hybrid shifted window-based multi-head self-attention (HSW-MSA) in place of the conventional shifted window-based multi-head self-attention (SW-MSA). This adjustment enables the model to more efficiently process areas of skin cancer overlap, capture finer details, and manage long-range dependencies, while maintaining memory usage and computational efficiency during training. Additionally, the study replaces the standard multi-layer perceptron (MLP) in the Swin Transformer with a SwiGLU-based MLP, an upgraded version of the gated linear unit (GLU) module, to achieve higher accuracy, faster training speeds, and better parameter efficiency. The modified Swin model-base was evaluated using the publicly accessible ISIC 2019 skin dataset with eight classes and was compared against popular convolutional neural networks (CNNs) and cutting-edge vision transformer (ViT) models. In an exhaustive assessment on the unseen test dataset, the proposed Swin-Base model demonstrated exceptional performance, achieving an accuracy of 89.36%, a recall of 85.13%, a precision of 88.22%, and an F1-score of 86.65%, surpassing all previously reported research and deep learning models documented in the literature.

Colon Cancer Detection of Colonoscopy Images Using CNN

Abstract: Colon cancer requires an early and precise diagnosis in order to be effectively treated because it is a common and potentially fatal condition. This study proposes a reliable framework for colon cancer detection through image preprocessing and transfer learning approaches. The research compares the performance of three popular convolutional neural networks (CNNs) for this task: VGG-15, ResNet-50, and AlexNet.The image preprocessing phase entails a number of critical operations, including image conversion, resizing, and contrast enhancement using histogram equalization. In order to guarantee that the neural networks receive high-quality inputs, these preprocessing techniques strive to improve the quality and consistency of the colon cancer dataset. Pre-trained models extract and classify features from big datasets using transfer learning. The preprocessed colon cancer dataset is used to fine-tune the chosen CNN architectures VGG-15, ResNet-50, and AlexNet and adapt them for this particular medical imaging purpose.The results of this study show that transfer learning has the potential for colon cancer detection and provides helpful direction for researchers and practitioners looking for the best deep-learning architectures for early cancer diagnosis. This study adds to continuing efforts to improve the precision and effectiveness of colon cancer detection techniques, ultimately leading to better patient outcomes.

Bird Species Detection Using Deep Learning

Bird species classification plays a crucial role in various ecological and conservation endeavors. Leveraging deep learning techniques has shown promise in automating this task, offering the potential for efficient and accurate species identification. This paper presents a comprehensive study on the application of the YOLOv5 architecture, a state-of-the-art object detection framework, for bird species classification from images. We explore the effectiveness of YOLOv5 in comparison to traditional convolutional neural networks (CNNs) for bird species identification. Our study involves extensive experimentation with different configurations of YOLOv5, including variations in model size, training strategies, and dataset preprocessing techniques. We evaluate the performance of the YOLOv5-based classification framework on diverse datasets, ranging from publicly available bird image datasets to custom datasets collected through field observations. Additionally, we compare the classification accuracy of YOLOv5 with other popular CNN architectures, such as ResNet and EfficientNet, to assess its efficacy in bird species recognition tasks. Furthermore, we investigate the transferability of pre-trained YOLOv5 models to different bird species datasets and examine the robustness of the models to variations in image quality, background clutter, and occlusions.Our experimental results demonstrate that YOLOv5 offers competitive performance in bird species classification tasks, achieving high accuracy and efficiency. We discuss the strengths and limitations of using YOLOv5 for this application and provide insights into potential avenues for future research.

Deep learning approaches for autonomous crack detection in concrete wall, brick deck and pavement

Detecting cracks is vital for inspecting and maintaining concrete structures, enabling early intervention and preventing potential damage. The advent of computer vision and image processing in civil engineering has ushered in deep learning-based semi-automatic/automatic techniques, replacing traditional visual inspections. These methods, driven by autonomous diagnosis, have applications across various sectors, fostering rapid progress in civil engineering. In this study, we present an approach that combines vision transformers and convolutional neural networks (CNN) for autonomously diagnosing cracks in bridges, roads, and walls. Performance enhancement was achieved through transfer learning, data augmentation, and optimized hyperparameters, utilizing popular CNN and ViT architectures. The proposed method was tested on the SDNET2018 dataset, comprising over 56,000 images. Experimental results demonstrated the approach's effectiveness, achieving high accuracy in detecting road cracks at 96.41%, wall cracks at 92.76%, and bridge cracks at 92.81%. These findings highlight the promising potential of deep learning in this field.

The Classification of VOCs Based on Sensor Images Using a Lightweight Neural Network for Lung Cancer Diagnosis.

The application of artificial intelligence to point-of-care testing (POCT) disease detection has become a hot research field, in which breath detection, which detects the patient's exhaled VOCs, combined with sensor arrays of convolutional neural network (CNN) algorithms as a new lung cancer detection is attracting more researchers' attention. However, the low accuracy, high-complexity computation and large number of parameters make the CNN algorithms difficult to transplant to the embedded system of POCT devices. A lightweight neural network (LTNet) in this work is proposed to deal with this problem, and meanwhile, achieve high-precision classification of acetone and ethanol gases, which are respiratory markers for lung cancer patients. Compared to currently popular lightweight CNN models, such as EfficientNet, LTNet has fewer parameters (32 K) and its training weight size is only 0.155 MB. LTNet achieved an overall classification accuracy of 99.06% and 99.14% in the own mixed gas dataset and the University of California (UCI) dataset, which are both higher than the scores of the six existing models, and it also offers the shortest training (844.38 s and 584.67 s) and inference times (23 s and 14 s) in the same validation sets. Compared to the existing CNN models, LTNet is more suitable for resource-limited POCT devices.