Deep Learning-based Approach Research Articles

A deep learning-based neural style transfer optimization approach

A deep learning-based neural style transfer optimization approach

Improving the Short-Range Precipitation Forecast of Numerical Weather Prediction through a Deep Learning-Based Mask Approach

Improving the Short-Range Precipitation Forecast of Numerical Weather Prediction through a Deep Learning-Based Mask Approach

A Survey of Galaxy Pairs in the SDSS Photometric Images based on Faster-RCNN

Galaxy pairs hold significant importance in understanding the evolution of galaxies, and the extensive search for a large sample of galaxy pairs is meaningful. In this article, we develop a deep learning-based approach for the search of galaxy pairs and conduct a comprehensive search on Sloan Digital Sky Survey (SDSS) images. In nine million photometric images, 17,965 physical galaxy pairs with spectral or photometric redshifts are detected. Four sets of results are provided, including physical pairs determined by two spectral redshifts, two photometric redshifts, one spectral redshift, and one photometric redshift, and visual irregular pairs that have no precise redshift information but can be inferred as physical galaxy pairs based on the morphological changes. Then their morphological and physical characteristics are explored, the redshifts of most targets are around 0.1, and as the redshift difference between two galaxies increases, the number of galaxy pairs gradually reduces. The distributions of star formation rate (SFR) are not the same for different morphologies of galaxy pairs, irregular pairs have higher SFR than the other three types, and statistics indicate that the SFR of galaxies depends on both nearby galaxies and internal properties. Color and stellar mass are also key properties of galaxies which can reflect the status of galaxy pairs. Compared to other surveys, a greater number of galaxy pair targets are detected, and this is also the first extensive detection of galaxy pairs in SDSS images using photometric redshifts. These galaxy pair samples can greatly aid in the study of galaxy evolution.

Deep learning-based prediction of 3-dimensional silver contact shapes enabling improved quality control in solar cell metallization

The industrial metallization of Si solar cells predominantly relies on screen printing, with silver as the preferred electrode material. However, the design of commercial screens often leads to suboptimal silver usage and increased electrical resistance due to print-related inhomogeneities like mesh marks, constrictions and spreading. Real-time monitoring of quality parameters during production has thus become increasingly critical. Current inline optical quality control systems usually only include 2D visualizations of the printed layout, which limits their effectiveness in quality control. Options that allow 3D measurements are usually slow, expensive, and therefore not worth considering in most cases. This research focuses on the development of a model that can estimate the three-dimensional shape of printed contact fingers from a single 2D image without the need of additional hardware using deep learning. Furthermore, a workflow for the generation of training data, which involves the creation of image pairs from a 2D microscope and a 3D confocal laser scanning microscope (CLSM) to accurately represent solar cell fingers, is presented. After model training, the predicted height maps are compared with the ground truth height maps, and the robustness of the model with respect to a paste variation and screen parameter variation is examined. The results confirm the feasibility and reliability of deep learning-based 3D shape estimation, extending its applicability to new, previously unseen data from screen-printed contact fingers. With a structural similarity index (SSIM) score of 0.76, a strong correlation between the estimated and ground truth height maps is established. In summary, our deep learning-based approach for height map estimation offers an effective and reliable solution for fast inline detection and analysis of the cross-sectional area of the printed contact fingers.

Automatic Optic Nerve Assessment From Transorbital Ultrasound Images: A Deep Learning-based Approach.

Transorbital Ultrasonography (TOS) is a promising imaging technology that can be used to characterize the structures of the optic nerve and the potential alterations that may occur in those structures as a result of an increase in intracranial pressure (ICP) or the presence of other disorders such as multiple sclerosis (MS) and hydrocephalus. In this paper, the primary objective is to develop a fully automated system that is capable of segmenting and calculating the diameters of structures that are associated with the optic nerve in TOS images. These structures include the optic nerve diameter sheath (ONSD) and the optic nerve diameter (OND). A fully convolutional neural network (FCN) model that has been pre-trained serves as the foundation for the segmentation method. The method that was developed was utilized to collect 464 different photographs from 110 different people, and it was accomplished with the assistance of four distinct pieces of apparatus. An examination was carried out to compare the outcomes of the automatic measurements with those of a manual operator. Both OND and ONSD have a typical inaccuracy of -0.12 0.32 mm and 0.14 0.58 mm, respectively, when compared to the operator. The Pearson correlation coefficient (PCC) for OND is 0.71, while the coefficient for ONSD is 0.64, showing that there is a positive link between the two measuring tools. A conclusion may be drawn that the technique that was developed is automatic, and the average error (AE) that was reached for the ONSD measurement is compatible with the ranges of inter-operator variability that have been discovered in the literature.

Deep Learning-Based Multifidelity Surrogate Modeling for High-Dimensional Reliability Prediction

Abstract Multifidelity surrogate modeling offers a cost-effective approach to reducing extensive evaluations of expensive physics-based simulations for reliability prediction. However, considering spatial uncertainties in multifidelity surrogate modeling remains extremely challenging due to the curse of dimensionality. To address this challenge, this paper introduces a deep learning-based multifidelity surrogate modeling approach that fuses multifidelity datasets for high-dimensional reliability analysis of complex structures. It first involves a heterogeneous dimension transformation approach to bridge the gap in terms of input format between the low-fidelity and high-fidelity domains. Then, an explainable deep convolutional dimension-reduction network (ConvDR) is proposed to effectively reduce the dimensionality of the structural reliability problems. To obtain a meaningful low-dimensional space, a new knowledge reasoning-based loss regularization mechanism is integrated with the covariance matrix adaptation evolution strategy (CMA-ES) to encourage an unbiased linear pattern in the latent space for reliability prediction. Then, the high-fidelity data can be utilized for bias modeling using Gaussian process (GP) regression. Finally, Monte Carlo simulation (MCS) is employed for the propagation of high-dimensional spatial uncertainties. Two structural examples are utilized to validate the effectiveness of the proposed method.

Accelerating multipool CEST MRI of Parkinson's disease using deep learning-based Z-spectral compressed sensing.

To develop a deep learning-based approach to reduce the scan time of multipool CEST MRI for Parkinson's disease (PD) while maintaining sufficient prediction accuracy. A deep learning approach based on a modified one-dimensional U-Net, termed Z-spectral compressed sensing (CS), was proposed to recover dense Z-spectra from sparse ones. The neural network was trained using simulated Z-spectra generated by the Bloch equation with various parameter settings. Its feasibility and effectiveness were validated through numerical simulations and in vivo rat brain experiments, compared with commonly used linear, pchip, and Lorentzian interpolation methods. The proposed method was applied to detect metabolism-related changes in the 6-hydroxydopamine PD model with multipool CEST MRI, including APT, CEST@2 ppm, nuclear Overhauser enhancement, direct saturation, and magnetization transfer, and the prediction performance was evaluated by area under the curve. The numerical simulations and in vivo rat-brain experiments demonstrated that the proposed method could yield superior fidelity in retrieving dense Z-spectra compared with existing methods. Significant differences were observed in APT, CEST@2 ppm, nuclear Overhauser enhancement, and direct saturation between the striatum regions of wild-type and PD models, whereas magnetization transfer exhibited no significant difference. Receiver operating characteristic analysis demonstrated that multipool CEST achieved better predictive performance compared with individual pools. Combined with Z-spectral CS, the scan time of multipool CEST MRI can be reduced to 33% without distinctly compromising prediction accuracy. The integration of Z-spectral CS with multipool CEST MRI can enhance the prediction accuracy of PD and maintain the scan time within a reasonable range.

Balancing accuracy and efficiency: A lightweight deep learning model for COVID 19 detection

The novel coronavirus disease in human was detected in Wuhan, China and spread rapidly across countries each time mutating itself as a new variant. It has impacted not only the patients but also medical infrastructure a lot. Several artificial intelligence-based systems for detection of covid-19 in chest images have been proposed. This study aims to develop a deep learning-based approach that efficiently detect Covid-19 using chest Computed Tomography (CT) images. In this work, we propose an efficient and lightweight deep-learning architecture based on the concept of a densely connected network for Covid-19 detection. The proposed architecture is designed by stacking dense modules in which each layer is directly connected to the other layers, with very fewer number of neurons at each layer. The dense connection is used to improve the learning capability of the proposed model and the fewer number of neurons make network computationally efficient. CT images for covid-19 detection are effectively used for the detection of Covid-19 patients. We train and test our model on SARS-CoV-2 CT-scan datasets. The experimental results are evaluated on accuracy, sensitivity and F1 score. The Adam optimizer with binary cross entropy loss function gives 97.2% accuracy with 0.98 sensitivity, 0.97 precision and 0.97 F1 score. The model achieves maximum accuracy of 88% at precision 0.97%, specificity 0.82 and F1 score 0.89 with Covid-CT dataset. The proposed model is compared with other state-of-the-art methods. The results of the proposed methods and its comparison shows acceptable level of performance on chest CT images.

Manager sentiment, policy uncertainty, ESG disclosure and firm performance: a large language model in corporate landscape

Purpose This study aims to examine the impact of manager sentiment on the firm performance (FP) of Indian-listed nonfinancial firms. Further, it endeavors to investigate the moderating role of economic policy uncertainty (EPU) and environment, social and governance (ESG) transparency in this relationship. Design/methodology/approach A noble manager sentiment is introduced using FinBERT, a bidirectional encoder representation from a transformers (BERT)-type large language model. Using this deep learning-based natural language processing approach implemented through a Python-generated algorithm, this study constructs a manager sentiment for each firm and year based on the management discussions and analysis (MD&A) report. This research uses the system GMM to examine how manager sentiment affects FP. Findings The empirical results suggest that managers’ optimistic outlook in MD&A corporate disclosure sections tends to present higher performance. This positive association remains consistent after several robustness checks – using propensity score matching and instrumental variable approach to address further endogeneity, using alternative proxies of manager sentiment and FP and conducting subsample analysis based on financial constraints. Furthermore, the authors observe that the relationship is more pronounced for ESG-disclosed firms and during the low EPU. Practical implications The results demonstrate that the manager sentiment strongly predicts FP. Thus, this study may provide valuable insight for academics, practitioners, investors, corporates and policymakers. Originality/value To the best of the authors’ knowledge, this is the first study to predict FP by using FinBERT-based managerial sentiment, particularly in an emerging market context.

A new deep learning-based approach for concrete crack identification and damage assessment

Concrete building structures are prone to cracking as they are subjected to environmental temperatures, freeze-thaw cycles, and other operational environmental factors. Failure to detect cracks in the key building structure at the early stage can result in serious accidents and associated economic losses. A new method using the SE-U-Net model based on a conditional generative adversarial network (CGAN) has been developed to identify small cracks in concrete structures in this paper. This proposed method was a pixel-level U-Net model based on a generative network, that was integrated the original convolutional layer with an attention mechanism, and an SE module in the jump connection section was added to improve the identifiability of the model. The discriminative network compared the generated images with real images using the PatchGAN model. Through the adversarial training of generator and discriminator, the performance of generator in crack image segmentation task is improved, and the trained generation network is used to segment cracks. In damage assessments, the crack skeleton was represented by the individual pixel width and recognized using the binary morphological crack skeleton method, in which the final length, area, and average width of the crack could be determined through the geometric correction index. The results showed that compared with other methods, the proposed method could better identify subtle pixel-level cracks, and the identification accuracy is 98.48%. These methods are of great significance for the identification of cracks and the damage assessment of concrete structures in practice.

Tractography from T1-weighted MRI: empirically exploring the clinical viability of streamline propagation without diffusion MRI

Abstract Over the last few decades, diffusion MRI (dMRI) streamline tractography has emerged as the dominant method for in vivo estimation of white matter (WM) pathways in the brain. One key limitation to this technique is that modern tractography implementations require high angular resolution diffusion imaging (HARDI). However, HARDI can be difficult to collect clinically, limiting the reach of tractography analyses to research cohorts and thus limiting many WM investigations to certain populations and pathologies. As such, a clinically viable tractography solution applicable to wider patient populations scanned as a part of routine care would be of key significance in broadening WM analyses to underfunded or rarer diseases and to the clinical setting. Such a solution would require the ability to perform arbitrary tractography analyses, use only clinical imaging for input, and be open source and widely accessible and implementable. Thus, here we evaluate our recently developed, containerized and open-source, T1-weighted (T1w) MRI-based deep learning model for streamline propagation. We empirically assess its performance against traditional dMRI-based and established atlas-based approaches in a healthy young population, an aging one, and in those with epilepsy, depression, and brain cancer. In the healthy young population, we find slightly increased error compared to traditional tractography with the deep learning model that falls within the bounds attributable to dMRI variability and is considerably less than the atlas-based approach. Further, seeking to replicate previously published dMRI tractography effects in the remaining cohorts as an initial assessment of clinical viability, we find this model successfully does so in some key cases—particularly in applications that rely on long-range streamlines including those not captured by the atlas-based approach—but importantly not all. These results suggest a deep learning-based approach to tractography with T1w MRI demonstrates promise within the limitations of our definition of clinical viability and especially over atlas-based approaches but requires refinement and more robust consideration of out-of-distribution effects prior to widespread clinical use. We also find these results raise additional questions regarding the differences in image content between dMRI and T1w MRI and their relationship to tractography. Further investigation of these questions will improve the field’s understanding of which features of the brain influence measured tractography effects.

What not to do in facial infrared thermographic measurements: A post data enhancement

The accuracy of infrared thermographic measurements depends on several factors, including movement of target. In this study, accuracy of nose tip temperatures obtained in a mental workload assessment using a thermal imaging camera were impacted by participants’ movement and camera zooming/panning. To correct these measurement errors, we compared manual facial landmark identification techniques using data labelling software with an automated deep learning-based approach utilised for facial landmark tracking and evaluated both against the built-in tracking features of the thermal camera, Thermal Spot Tracking. Using the Manual Thermal Landmark Annotation measurements as the ground truth, our results show that the Automated Facial Feature Tracking approach, which is the AI based approach performed better than the Thermal Spot Tracking as it matched comparatively more spatial coordinates and temperature datapoints as well as showed comparatively lower mean relative error. The study highlights the potential of AI in enhancing the accuracy of thermographic measurements, particularly in applications involving facial temperature analysis.

Hybrid deep learning-based identification of microseismic events in TBM tunnelling

For TBM’s safe and efficient tunnelling, the microseismic monitoring technique has been widely applied in rockburst warning. However, useful microseismic events are often mixed with noisy events, which seriously affects warning accuracy. To this end, this study proposes several hybrid deep learning-based microseismic identification approaches in TBM tunnelling, where recurrent neural networks directly treat monitored microseismic events as time series and avoid complex feature pre-extraction, and grey wolf optimization and attention mechanism are embedded to optimize model hyper-parameters and recognize value information. Additionally, the learning curve and early-stopping strategy are also integrated to reduce overfitting and underfitting risks. Results in a real TBM-excavated water conveyance tunnel indicate that the bidirectional long short-term memory network model combined with grey wolf optimization and attention mechanism achieves the greatest identification performance among discussed models and provides an effective support for intelligent identification of microseismic events under TBM tunnelling environments.

A Boundary-Enhanced Decouple Fusion Segmentation Network for Diagnosis of Adenomatous Polyps.

Adenomatous polyps, a common premalignant lesion, are often classified into villous adenoma (VA) and tubular adenoma (TA). VA has a higher risk of malignancy, whereas TA typically grows slowly and has a lower likelihood of cancerous transformation. Accurate classification is essential for tailored treatment. In this study, we develop a deep learning-based approach for the localization and classification of adenomatous polyps using endoscopic images. Specifically, a pre-trained EGE-UNet is first adopted to extract regions of interest from original images. Multi-level feature maps are then extracted by the feature extraction pipeline (FEP). The deep-level features are fed into the Pyramid Pooling Module (PPM) to capture global contextual information, and the squeeze body edge (SBE) module is then used to decouple the body and edge parts of features, enabling separate analysis of their distinct characteristics. The Group Aggregation Bridge (GAB) and Boundary Enhancement Module (BEM) are then applied to enhance the body features and edge features, respectively, emphasizing their structural and morphological characteristics. By combining the features of the body and edge parts, the final output can be obtained. Experiments show the proposed method achieved promising results on two private datasets. For adenoma vs. non-adenoma classification, It achieved a mIoU of 91.41%, mPA of 96.33%, mHD of 11.63, and mASD of 2.33. For adenoma subclassification (non-adenomas vs. villous adenomas vs. tubular adenomas), it achieved a mIoU of 91.21%, mPA of 94.83%, mHD of 13.75, and mASD of 2.56. These results demonstrate the potential of our approach for precise adenomatous polyp classification.

DRL-Tomo: a deep reinforcement learning-based approach to augmented data generation for network tomography

Abstract Accurate and current comprehension of network status is crucial for efficient network management. Nevertheless, direct network measurement strategies entail substantial traffic overhead and demand intricate coordination among network entities, making them impractical. Network tomography, an indirect measurement approach, utilizes insights garnered from measured parts to deduce characteristics of the entire network. Past studies frequently depend on acquiring challenging-to-access information, such as the complete network topology or support from specialized protocols. Unfortunately, these constraints pose challenges in non-cooperative scenarios where obtaining such information is difficult. Recent endeavors pursue emancipating tomography from dependence on copious information, striving to predict unmeasured path performance using limited data. Nevertheless, the disparity between the measured data and actual performance has hindered the accuracy. In response, we introduce an innovative tomography framework named DRL-Tomo, designed to alleviate potential biases. DRL-Tomo initiates by generating augmented data through deep reinforcement learning, gradually approximating the genuine performance of unmeasured paths. Subsequently, a neural network model is trained using this augmented data, enabling precise inferences. Our experiments, encompassing both real-world and synthetic datasets, vividly demonstrate DRL-Tomo’s remarkable enhancement. Specifically, it achieves a substantial 10%–67% improvement in path delay prediction and an impressive 30%–98% enhancement in path loss rate prediction.

Enhancing vehicle detection in intelligent transportation systems via autonomous UAV platform and YOLOv8 integration

This study highlights the evolving landscape of object detection methodologies, emphasizing the superiority of deep learning-based approaches over traditional methods. Particularly in intelligent transportation systems-related applications requiring robust image processing techniques, such as vehicle identification, localization, tracking, and counting within traffic scenarios, deep learning has gained substantial traction. The YOLO algorithm, in its various iterations, has emerged as a popular choice for such tasks, with YOLOv5 garnering significant attention. However, a more recent iteration, YOLOv8, was introduced in early 2023, ushering in a new phase of exploration and potential innovation in the field of object detection. Consequently, due to its recent emergence, the number of studies on YOLOv8 is extremely limited, and an application in the field of Intelligent Transportation Systems (ITS) has not yet found its place in the existing literature. In light of this gap, this study makes a noteworthy contribution by delving into vehicle detection using the YOLOv8 algorithm. Specifically, the focus is on targeting aerial images acquired through a modified autonomous UAV, representing a unique avenue for the application of this cutting-edge algorithm in a practical context. The dataset employed for training and testing the algorithm was curated from a diverse collection of traffic images captured during UAV missions. In a strategic effort to enhance the variability of vehicle images, the study systematically manipulated flight patterns, altitudes, orientations, and camera angles through a custom-designed and programmed drone. This deliberate approach aimed to bolster the algorithm's adaptability across a wide spectrum of scenarios, ultimately enhancing its generalization capabilities. To evaluate the performance of the algorithm, a comprehensive comparative analysis was conducted, focusing on the YOLOv8n and YOLOv8x submodels within the YOLOv8 series. These submodels were subjected to rigorous testing across diverse lighting and environmental conditions using the dataset. Through tests, it was observed that YOLOv8n achieved an average precision of 0.83 and a recall of 0.79, whereas YOLOv8x attained an average precision of 0.96 and a recall of 0.89. Furthermore, YOLOv8x also outperformed YOLOv8n in terms of F1 score and mAP, achieving values of 0.87 and 0.83 respectively, compared to YOLOv8n's 0.81 and 0.79. These outcomes of the evaluation illuminated the relative strengths and weaknesses of YOLOv8n and YOLOv8x, leading to the conclusion that YOLOv8n is well-suited for real-time ITS applications, while YOLOv8x exhibits superior detection capabilities.

Enhancing image watermarking efficiency through advanced deep learning: A novel approach with modified Res-Ception blocks and practical applications in modern scenarios

In recent years, digital media has become a prevalent form of information exchange, necessitating robust measures for privacy, data safety, and copyright protection. Image watermarking is one of the tools that can be employed for these tasks. Research on blind image watermarking techniques is lacking, and there is a gap between theoretical findings and practical implementations to address contemporary issues such as photo metadata privacy, AI generated image recognition, etc. The proposed research introduces a revolutionary deep learning-based image watermarking approach in order to address these issues. This scheme is designed to overcome the contemporary challenges related to image watermarking by being blind, efficient and robust. It also has an improved PSNR and beats state-of-the-art in various attacks. The model outperforms existing deep learning frameworks in terms of fidelity. All these are achieved by creating a modified and much more efficient form of Inception Resnet block, coined ResCeption Block, for use in effective learning of watermarking networks. The research also presents an application of watermarking in the identification of AI generated images and proposes a novel and robust algorithm for metadata encryption of digital images and enforce digital media security. Rigorous testing has been performed on benchmark datasets, and comparisons have been done against state-of-the-art and a PSNR of 43.12 dB has been achieved.

Atmospheric Turbulence Phase Reconstruction via Deep Learning Wavefront Sensing.

The fast and accurate reconstruction of the turbulence phase is crucial for compensating atmospheric disturbances in free-space coherent optical communication. Traditional methods suffer from slow convergence and inadequate phase reconstruction accuracy. This paper introduces a deep learning-based approach for atmospheric turbulence phase reconstruction, utilizing light intensity images affected by turbulence as the basis for feature extraction. The method employs extensive light intensity-phase samples across varying turbulence intensities for training, enabling phase reconstruction from light intensity images. The trained U-Net model reconstructs phases for strong, medium, and weak turbulence with an average processing time of 0.14 s. Simulation outcomes indicate an average loss function value of 0.00027 post-convergence, with a mean squared error of 0.0003 for individual turbulence reconstructions. Experimental validation yields a mean square error of 0.0007 for single turbulence reconstruction. The proposed method demonstrates rapid convergence, robust performance, and strong generalization, offering a novel solution for atmospheric disturbance correction in free-space coherent optical communication.

Assessment of image quality and diagnostic accuracy for cervical spondylosis using T2w-STIR sequence with a deep learning-based reconstruction approach.

To investigate potential of enhancing image quality, maintaining interobserver consensus, and elevating disease diagnostic efficacy through the implementation of deep learning-based reconstruction (DLR) processing in 3.0T cervical spine fast magnetic resonance imaging (MRI) images, compared with conventional images. The 3.0T cervical spine MRI images of 71 volunteers were categorized into two groups: sagittal T2-weighted short T1 inversion recovery without DLR (Sag T2w-STIR) and with DLR (Sag T2w-STIR-DLR). The assessment covered artifacts, perceptual signal-to-noise ratio, clearness of tissue interfaces, fat suppression, overall image quality, and the delineation of spinal cord, vertebrae, discs, dopamine, and joints. Spanning canal stenosis, neural foraminal stenosis, herniated discs, annular fissures, hypertrophy of the ligamentum flavum or vertebral facet joints, and intervertebral disc degeneration were evaluated by three impartial readers. Sag T2w-STIR-DLR images exhibited markedly superior performance across quality indicators (median = 4 or 5) compared to Sag T2w-STIR sequences (median = 3 or 4) (p < 0.001). No statistically significant differences were observed between the two sequences in terms of diagnosis and grading (p > 0.05). The interobserver agreement for Sag T2w-STIR-DLR images (0.604-0.931) was higher than the other (0.545-0.853), Sag T2w-STIR-DLR (0.747-1.000) demonstrated increased concordance between reader 1 and reader 3 in comparison to Sag T2w-STIR (0.508-1.000). Acquisition time diminished from 364 to 197s through the DLR scheme. Our investigation establishes that 3.0T fast MRI images subjected to DLR processing present heightened image quality, bolstered diagnostic performance, and reduced scanning durations for cervical spine MRI compared with conventional sequences.

A Smart Visual Sensor for Smoke Detection Based on Deep Neural Networks.

The automatic detection of smoke by analyzing the video stream acquired by traditional surveillance cameras is becoming a more and more interesting problem for the scientific community thanks to the necessity to prevent fires at the very early stages. The adoption of a smart visual sensor, namely a computer vision algorithm running in real time, allows one to overcome the limitations of standard physical sensors. Nevertheless, this is a very challenging problem, due to the strong similarity of the smoke with other environmental elements like clouds, fog and dust. In addition to this challenge, data available for training deep neural networks is limited and not fully representative of real environments. Within this context, in this paper we propose a new method for smoke detection based on the combination of motion and appearance analysis with a modern convolutional neural network (CNN). Moreover, we propose a new dataset, called the MIVIA Smoke Detection Dataset (MIVIA-SDD), publicly available for research purposes; it consists of 129 videos covering about 28 h of recordings. The proposed hybrid method, trained and evaluated on the proposed dataset, demonstrated to be very effective by achieving a 94% smoke recognition rate and, at the same time, a substantially lower false positive rate if compared with fully deep learning-based approaches (14% vs. 100%). Therefore, the proposed combination of motion and appearance analysis with deep learning CNNs can be further investigated to improve the precision of fire detection approaches.