Heavy Fog Research Articles

An in-vehicle real-time infrared object detection system based on deep learning with resource-constrained hardware

Advanced driver assistance systems primarily rely on visible images for information. However, in low-visibility weather conditions, such as heavy rain or fog, visible images struggle to capture road conditions accurately. In contrast, infrared (IR) images can overcome this limitation, providing reliable information regardless of external lighting. Addressing this problem, we propose an in-vehicle IR object detection system. We optimize the you only look once (YOLO) v4 object detection algorithm by replacing its original backbone with MobileNetV3, a lightweight feature extraction network, resulting in the MobileNetV3-YOLOv4 model. Furthermore, we replace traditional pre-processing methods with an Image Enhancement Conditional Generative Adversarial Network inversion algorithm to enhance the pre-processing of the input IR images. Finally, we deploy the model on the Jetson Nano, an edge device with constrained hardware resources. Our proposed method achieves an 82.7% mean Average Precision and a frame rate of 55.9 frames per second on the FLIR dataset, surpassing state-of-the-art methods. The experimental results confirm that our approach provides outstanding real-time detection performance while maintaining high precision.

Object classification through heterogeneous fog with a fast data-driven algorithm using a low-cost single-photon avalanche diode array

This study presents a framework for classifying a wooden mannequin’s poses using a single-photon avalanche diode (SPAD) array in dynamic and heterogeneous fog conditions. The target and fog generator are situated within an enclosed fog chamber. Training datasets are continuously collected by configuring the temporal and spatial resolutions on the sensor's firmware, utilizing a low-cost SPAD array sensor priced below $5, consisting of an embedded SPAD array and diffused VCSEL laser. An extreme learning machine (ELM) is trained for rapid pose classification, as a benchmark against CNN. We quantitatively justify the selection of nodes in the hidden layer to balance the computing speed and accuracy. Results demonstrate that ELM can accurately classify mannequin poses when obscured by dynamic heavy fog to 35 cm away from the sensor, enabling real-time applications in consumer electronics. The proposed ELM achieves 90.65% and 89.58% accuracy in training and testing, respectively. Additionally, we demonstrate the robustness of both ELM and CNN as the fog density increases. Our study also discusses the sensor’s current optical limitations and lays the groundwork for future advancements in sensor technology.

Improved DeepSORT-Based Object Tracking in Foggy Weather for AVs Using Sematic Labels and Fused Appearance Feature Network.

The presence of fog in the background can prevent small and distant objects from being detected, let alone tracked. Under safety-critical conditions, multi-object tracking models require faster tracking speed while maintaining high object-tracking accuracy. The original DeepSORT algorithm used YOLOv4 for the detection phase and a simple neural network for the deep appearance descriptor. Consequently, the feature map generated loses relevant details about the track being matched with a given detection in fog. Targets with a high degree of appearance similarity on the detection frame are more likely to be mismatched, resulting in identity switches or track failures in heavy fog. We propose an improved multi-object tracking model based on the DeepSORT algorithm to improve tracking accuracy and speed under foggy weather conditions. First, we employed our camera-radar fusion network (CR-YOLOnet) in the detection phase for faster and more accurate object detection. We proposed an appearance feature network to replace the basic convolutional neural network. We incorporated GhostNet to take the place of the traditional convolutional layers to generate more features and reduce computational complexities and costs. We adopted a segmentation module and fed the semantic labels of the corresponding input frame to add rich semantic information to the low-level appearance feature maps. Our proposed method outperformed YOLOv5 + DeepSORT with a 35.15% increase in multi-object tracking accuracy, a 32.65% increase in multi-object tracking precision, a speed increase by 37.56%, and identity switches decreased by 46.81%.

Infrared and visible image fusion and detection based on interactive training strategy and feature filter extraction module

The infrared and visible image fusion-detection technique aims to expand the observation wavelength range to overcome harsh environmental conditions such as low light, smoke, heavy fog, etc., and improve the reliability and robustness of target detection. However, the current fusion-detection method suffers from problems of the joint optimization of the fusion-detection model and irrelevant feature interference. To address these problems, we propose an infrared and visible image fusion-detection network based on an interactive training strategy (ITS) and a feature filter extraction module (FFEM). The ITS constructs the closed-loop optimization for the fusion-detection method, which can mutually enhance the performance of the fusion and detection model through mutual supervision. The FFEM is designed to acquire detection-conducive features from visible and infrared images. Specifically, we define filtering weights based on the covariance values of features and use these weights to enhance relevant features and soft redundant and irrelevant features (e.g., noise). Quantitative and qualitative experimental results demonstrate that the proposed method achieves competitive performance in target detection accuracy compared with other state-of-the-art fusion-detection methods.

Legibility of variable message signs on foggy highway: Effect of text color and spacing

Variable message signs are frequently utilized as highway communication facilities to inform vehicles about the state of the roads and the weather. However, standards and design specifications for variable message signs are unclear. This study examines the effect of the own factors and environmental factors on the legibility of variable message signs in order to optimize the design of variable message signs on highways. Three font colors (green, red, and yellow) and three levels of word spacing (0.1H, 0.3H, and 0.5H) were chosen for the variable message signs. For the environmental factors, three different weather types were represented according to visibility, namely sunny (visibility 10,000 m), light fog (visibility = 1000), and heavy fog (visibility = 200). The legibility time and distance were recorded for 23 participants, and then the legibility of the variable message signs was analyzed. The results showed that own factors and environmental factors influenced the legibility of the variable message signs. Experiment 1 results showed that the legibility of the variable message signs was the best when the text was yellow in light fog weather conditions. Experiment 2 results showed that the legibility of the variable message signs was better than other experimental scenarios in light fog conditions with a 0.3H word spacing. The findings of this study will be used by relevant agencies to establish design guidelines for variable message signs.

Design and performance evaluation of a 350 m free space optical communications link for pico-macrocell backhauling

Fibreless optics or free space optical communications (FSOC) has been at the forefront of many academic research in telecommunications due to its numerous benefits of large spectrum, high-speed data transmission, security, low transmit power, unlicensed spectrum and non-interfering links. Among the technical challenges of dense deployment of small cells in heterogeneous networks (HetNet) is a flexible and cost-effective backhaul link. This paper proposes, designs, simulates and evaluates the performance of a 350 m FSOC link under different atmospheric impairments for picocell to macrocell backhauling applications. The performance of the FSOC link is assessed by evaluating bit error rate (BER), eye diagram and quality factor (Q-factor). Results obtained recommend the FSOC link deployment for pico-macrocell backhauling under the weather conditions of clear sky with/without turbulence, heavy rain, heavy haze, heavy fog and wet snow.

Mitigation of atmospheric losses using dual‐polarized 16‐QAM MIMO technique in free‐space optical system in various weather conditions

SummaryThe technique to overcome the challenge while implementing the fiber optical link is investigated using an alternative optical link called a free‐space optical (FSO) link with advanced modulation and signal processing techniques. A dual‐polarized 16‐quadrature amplitude‐modulated signal is transmitted over single and multi‐FSO channels under various atmospheric attenuation conditions. High bandwidth and data rates (up to 10 Gbps), unlicensed spectrum (above 300 GHz), and simplicity of deployment are significant benefits of the FSO system. However, the weather conditions of haze, rain, and fog cause signal attenuation when the optical carrier wave is transmitted over free space, and atmospheric turbulence is a challenge to FSO implementation, which degrades the transmission performance of optical signals for long‐distance communication. For 120 Gbps data transmission, the performance of a multi‐input and multi‐output (MIMO) FSO system is evaluated by observing the bit error rate (BER) and received signal constellation diagram. Optical signal‐to‐noise ratio (OSNR) parameters were iterated to determine the optimal value for back‐to‐back, single‐input, single‐output (SISO), and MIMO systems. Moreover, we examined the proposed system's output with recent articles, including MIMO, multiplexing, and various modulation techniques. The comparison of results shows that the suggested MIMO‐FSO model has enhanced the performance with the lower BER as less than 3.8 × 10−3 (log of BER is −2.42) forward error correction (FEC) limit for 18.7 km link range in clear air with 0.54 dB/km attenuation, 4.8, 4.5, 1.35, and 0.57 km under haze, light fog, medium, and heavy fog with 4.6, 6.9, 28.9, and 75 dB/km attenuations.

AIE-YOLO: Effective object detection method in extreme driving scenarios via adaptive image enhancement.

The widespread research and implementation of visual object detection technology have significantly transformed the autonomous driving industry. Autonomous driving relies heavily on visual sensors to perceive and analyze the environment. However, under extreme weather conditions, such as heavy rain, fog, or low light, these sensors may encounter disruptions, resulting in decreased image quality and reduced detection accuracy, thereby increasing the risk for autonomous driving. To address these challenges, we propose adaptive image enhancement (AIE)-YOLO, a novel object detection method to enhance road object detection accuracy under extreme weather conditions. To tackle the issue of image quality degradation in extreme weather, we designed an improved adaptive image enhancement module. This module dynamically adjusts the pixel features of road images based on different scene conditions, thereby enhancing object visibility and suppressing irrelevant background interference. Additionally, we introduce a spatial feature extraction module to adaptively enhance the model's spatial modeling capability under complex backgrounds. Furthermore, a channel feature extraction module is designed to adaptively enhance the model's representation and generalization abilities. Due to the difficulty in acquiring real-world data for various extreme weather conditions, we constructed a novel benchmark dataset named extreme weather simulation-rare object dataset. This dataset comprises ten types of simulated extreme weather scenarios and is built upon a publicly available rare object detection dataset. Extensive experiments conducted on the extreme weather simulation-rare object dataset demonstrate that AIE-YOLO outperforms existing state-of-the-art methods, achieving excellent detection performance under extreme weather conditions.

ABCNet: A comprehensive highway visibility prediction model based on attention, Bi-LSTM and CNN.

Meteorological disasters along highways significantly reduce road traffic efficiency. Low visibility caused by heavy fog is a severe meteorological disaster that greatly increases highway traffic accidents. Accurately predicting highway visibility and taking timely response measures can reduce the impact of meteorological disasters and improve traffic safety. We proposed an Attention-based BiLSTM-CNN (ABCNet) model, which synergized attention mechanisms with BiLSTM and CNN technologies to forecast atmospheric visibility more accurately. First, the Bi-LSTM module processed information both forward and backward, capturing intricate temporal dependencies in the model. Second, the multi-head attention mechanism following the Bi-LSTM distilled and prioritized salient features from multiple aspects of the sequence data. Third, the CNN module recognized local spatial features, and a singular attention mechanism refined the feature map after the CNN module, further enhancing the model's accuracy and predictive capability. Experiments showed that the model was accurate, effective, and significantly advanced compared to conventional models. It could fully extract the spatiotemporal characteristics of meteorological elements. The model was integrated into practical systems with positive results. Additionally, this study provides a self-collected meteorological dataset for highways in high-altitude mountainous areas.

Winter avifauna habitat types of urban agglomerations: comparison by qualitative and quantitative indicators

Background. The winter avifauna of urban agglomerations is a rather dynamic object of research, whose indicators differ across cities and years of observation. Depending on the species characteristics and the level of anthropogenization, birds in the territory of cities choose different types of habitats, which also requires analysis. Materials and Methods. Research on the species composition of wintering birds on the territories of Zhydachiv and Lviv urban agglomerations was carried out in winter periods of 2020–2023 by means of field trips using the method of point records (Bibby et al., 2000). In order to count small species of Passeriformes accurately, a slightly modified version of this method was used. In this case, observations were carried out in a double radius. Birds were recorded separately in R1 – up to 25 m and R2 – from 25 to 50 m. The optimal time for counting at one point is 8–10 minutes. The counting area corresponds to a circle with a known radius. Research was carried out mainly in morning hours – from 8:00 to 11:00 in the case of absence of heavy precipitation, fog or strong wind. Results and Discussion. After comparing the winter avifauna of urban agglo­merations of the regional (Lviv) and district (Zhydachiv) levels, which differ in the level of environmental transformation, we established certain differences. A total of 48 species of wintering birds have been recorded, of which the Сommon Goldeneye (Bucephala clangula) and the Ural Owl (Strix uralensis) are protected by the Red Data Book of Ukraine. Conclusion. For each of the agglomerations, dominant species of birds have been determined. Admittedly, birds in the territory of cities seek to occupy a territory with conditions most similar to the natural environment. Additionally, we attempted to understand the patterns of distribution of wintering species in the study area and the reasons for such placement.

On the performance of FSO communication system with WDM and MIMO structure under different turbulent atmospheric conditions

Abstract Free space optical (FSO) communication link is highly weather dependent as the signal passes through the atmospheric channel. The main impairment is atmospheric turbulence, which introduces fading and breakdown the system performance. The performance of FSO link is evaluated in terms of attenuation and link length under different weather conditions. Taking a step further toward the improvement of FSO communication system, wavelength division multiplexing (WDM) is widely used in optical fiber in order to exploit the capacity of a fiber channel more efficiently. Additionally, multiple input multiple output (MIMO) mechanism represents one of the atmospheric mitigation techniques. Thus, integration of MIMO and WDM will provide astonishing results in FSO communication systems. This paper particularly focuses on analyzing the performance of FSO single input single output (FSO-SISO), FSO-WDM, FSO-MIMO, and FSO-WDM-MIMO systems considering different weather conditions. Performance of the different systems has also been compared in terms of received signal power, the quality factor, the bit error rate, and the maximum separation between the transmitter and the receiver for a given received signal level and for various atmospheric conditions. Our results show that increasing the transmitted power and receiver aperture diameter will improve the performance of the system under heavy fog and dense fog.

Enhancing fog collection by optimizing wettability combination and fork-row collector arrangement: light and heavy fog

Fog collection is essential to alleviate water scarcity in arid areas. However, the vast majority of existing fog collectors only work effectively in heavy fog. To broaden the fog concentration range for efficient work, an optimization strategy based on wettability combination and fork-row arrangement is proposed in this work. Single-layer experiment results show that a hydrophilic (HL) surface with high deposition capacity collects droplets at a faster rate in light fog (25–80 g h−1), while the collection rate of a hydrophobic (HB) surface with high drainage capacity is higher in heavy fog (220–500 g h−1). Double-layer experiment results show that in light fog, the best combination of double-layer collection electrodes is HL–HL, while HB–HL performs best in heavy fog. A 35% improvement in collection rate can be obtained simply by changing the arrangement from smooth-row (S) to fork-row (F), which is attributed to the increased effective collision area of droplets. In our series of experiments, at 50 g h−1, the collection rate of double-layer combination HL–HL(F) is 56.7% higher than that of single-layer HL. In particular, the collection rate of HB–HL(F) reaches 1434.7 mg cm−2 h−1 at 500 g h−1. Such a good performance is attributed to the force imbalance of hanging droplets caused by wettability differences, which tends to transport small droplets on HB towards HL directionally, resulting in a rapid droplet slippage. Therefore, the HB–HL accelerates drainage and refreshes capture points. Furthermore, fog collection performance is also influenced by layer spacing, which has an optimal distance. These findings provide a promising method for practical applications of fog collectors in a wide range of fog flow quantities, enhancing adaptability to variable environments.

Analysis of the different influence between initial/boundary and physical perturbation during ensemble forecast of fog

To explore the different effects of initial/boundary condition (ICBC) and physical perturbation during fog ensemble forecast, ensemble forecast experiments are done for a heavy fog episode from December 31, 2016 to January 2, 2017. Three ensemble schemes [ICBC, multi-physics (MPY), and a combination of ICBC and MPY (COM)] were compared. Their forecast performances are analyzed in detail and compared with the reference deterministic forecast. The results show that all ensemble schemes, especially the COM, are able to noticeably improve fog prediction. The TS score of ensemble-based fog forecast with 50% probability threshold is higher than that of the control deterministic prediction by ~ 26%. Compared with the ICBC scheme, the MPY scheme can produce a larger ensemble spread and has more skill in fog and the near-surface variables forecast. When ICBC and MPY are combined (the COM scheme), the ensemble spread is enhanced and the prediction performance is also further improved. The sensitivity experiments of different physical parameterization schemes are also analyzed among microphysics, planetary boundary layer, and land surface. The fog forecast is found to be most sensitive to the land surface scheme, followed by planetary boundary layer scheme, and the least sensitive to microphysics.

Failure Characterization of Silicone Rubber in Corrosive Environments Based on Leakage Current Characteristics

Due to the growing usage of composite insulators in transmission and distribution networks, the development of composite insulators has reached the stage of systematic reliability study. This paper discusses the failure properties of HTV silicone rubber in different energized corrosive environments and explores the characteristics that can characterize insulation failure. The findings demonstrate that the leakage current characteristics (the maximum values, the number of characteristic pulses, and cumulative charge) can effectively characterize the surface state of silicone rubber in corrosive environments. The flashover tests during the withstand voltage experiment accelerated the electrical performance degradation of silicone rubber and increased the leakage current. The three leakage current characteristics increase with experimental time, applied voltage, and fog conductivity. The occurrence of flashover is related to the fluctuation range of leakage current. The three leakage current characteristics in the alkali fog with the same conductivity and applied voltage are significantly higher than those in salt and acid fog. A failure characterization method of silicone rubber based on the proportion of pulse number and cumulative discharge is proposed in this paper. The findings are helpful in forecasting pollution flashovers and insulation failure in heavy fog, coastal, and industrial pollution areas.

An improved YOLO v4 used for grape detection in unstructured environment.

Visual recognition is the most critical function of a harvesting robot, and the accuracy of the harvesting action is based on the performance of visual recognition. However, unstructured environment, such as severe occlusion, fruits overlap, illumination changes, complex backgrounds, and even heavy fog weather, pose series of serious challenges to the detection accuracy of the recognition algorithm. Hence, this paper proposes an improved YOLO v4 model, called YOLO v4+, to cope with the challenges brought by unstructured environment. The output of each Resblock_body in the backbone is processed using a simple, parameterless attention mechanism for full dimensional refinement of extracted features. Further, in order to alleviate the problem of feature information loss, a multi scale feature fusion module with fusion weight and jump connection structure was pro-posed. In addition, the focal loss function is adopted and the hyperparameters α, γ are adjusted to 0.75 and 2. The experimental results show that the average precision of the YOLO v4+ model is 94.25% and the F1 score is 93%, which is 3.35% and 3% higher than the original YOLO v4 respectively. Compared with several state-of-the-art detection models, YOLO v4+ not only has the highest comprehensive ability, but also has better generalization ability. Selecting the corresponding augmentation method for specific working condition can greatly improve the model detection accuracy. Applying the proposed method to harvesting robots may enhance the applicability and robustness of the robotic system.

Research on the Detection Method of Coal Mine Roadway Bolt Mesh Based on Improved YOLOv7

Aiming at the environment of low illumination, high dust, and heavy water fog in coal mine driving face and the problems of occlusion, coincidence, and irregularity of bolt mesh laid on coal wall, a YOLOv7 bolt mesh-detection algorithm combining the image enhancement and convolutional block attention module is proposed. First, the image brightness is enhanced by a hyperbolic mapping transform-based image enhancement algorithm, and the image is defogged by a dark channel-based image defogging algorithm. Second, by introducing a convolutional block attention model in the YOLOv7 detection network, the significance of bolt mesh targets in the image is improved, and its feature expression ability in the detection network is enhanced. Meanwhile, the original activation function ReLU in the convolutional layer Conv of the YOLOv7 network is replaced by LeakyReLU so that the activation function has stronger nonlinear expression capability, which enhances the feature extraction performance of the network and thus improves the detection accuracy. Finally, the training and testing samples were prepared using the actual video of the drilling and bolting operation, and the proposed algorithm is compared with five classical target detection algorithms. The experimental results show that the proposed algorithm can be better applied to the low illumination, high dust environment, and irregular shape on the detection accuracy of coal mine roadway bolt mesh, and the average detection accuracy of the image can reach 95.4% with an average detection time of 0.0392 s.

Fog Density Evaluation by Combining Image Grayscale Entropy and Directional Entropy

The fog density level, as one of the indicators of weather conditions, will affect the management decisions of transportation management agencies. This paper proposes an image-based method to estimate fog density levels to improve the accuracy and efficiency of analyzing fine meteorological conditions and validating fog density predictions. The method involves two types of image entropy: a two-dimensional directional entropy derived from four-direction Sobel operators, and a combined entropy that integrates the image directional entropy and grayscale entropy. For evaluating the performance of the proposed method, an image test set and an image training set are constructed; and each image is labeled as heavy fog, moderate fog, light fog, or fog-free according to the fog density level of the image based on a user study. Using our method, the average accuracy rates of image fog level estimation were 77.27% and 79.39% on the training set using the five-fold cross-validation and the test set, respectively. Our experimental results demonstrate the effectiveness of the proposed combined entropy for image-based fog density level estimation.

FSO communication system for high-speed trains under varying visibility conditions

High-speed train systems are rapidly growing in complexity and coverage, demanding efficient high-speed internet services. Existing radio frequency technology struggles to meet these demands due to handover mechanisms and limited connectivity times. Furthermore, current free-space optics (FSO) techniques require a large number of base stations (BSs), resulting in high costs. This study presents a novel high-speed train communication technique, using FSO, that significantly reduces the number of required BSs while maintaining sufficient coverage and data speeds. The proposed technique can be adapted for fixed or variable representations of transmitted and received power, providing a fixed or variable data rate. The research also investigates the impact of divergence angle on transmitted and received power. By employing the proposed communication technique, a system with a fixed data rate, such as 1.25 Gbps, can be achieved. The results demonstrate the potential of this approach in high-speed train applications, reducing the number of BSs from 113, as reported in the best existing literature, to less than 10. Weather conditions, including clear, haze, thin fog, light fog, and heavy fog, are considered when calculating the required BS locations to achieve a target data rate. Additionally, this work introduces a new handover mechanism and mathematical formulation, providing a more comprehensive solution for high-speed train communication systems. The simulations and evaluations conducted in this study are based on parameters and equations from previous works to ensure a fair comparison.

Image transmission performance analysis through free space optical communication link using coherent QPSK modulation under various environmental conditions

AbstractFree space optics (FSO) technology has been used recently in many applications due to its various advantages such as high bandwidth, easy deployment, high security, free installation license and immunity to electromagnetic interference. However, this type of communication is vulnerable to meteorological conditions such as rain, haze and fog which significantly degrade the system's efficiency. To enhance the effectiveness of the FSO systems, Quadrature Phase Shift Keying (QPSK) modulation based on coherent detection constitutes a promising technique for achieving high spectral efficiency and long‐distance FSO links with the characteristics of low‐cost implementation and simple design. This work aims to investigate the performance of FSO communication links on image transmission under the presence of atmospheric attenuations. A comparative study has been made using QPSK modulation scheme and on–off‐keying (OOK) modulation scheme to estimate the maximum range that was used to recover an RGB transmitted image over an FSO link under different meteorological weathers. The obtained results clearly show that the quality of the received image has degraded significantly during transmission through free space yielding to limit the propagation range. Moreover, the proposed FSO communication system performs well when using QPSK modulation rather than OOK modulation in different environmental conditions. Thus, to improve the quality of the received image under adverse fog weather condition, high power and small beam divergence angle have been suggested to increase system's performance. The proposed solution increases the communication range of the FSO link by 230 m and 100 m for moderate and heavy fog, respectively.

Performance investigations on integrated MMF/FSO transmission enabled by OAM beams

This paper introduces a new hybrid system for high-bandwidth and high-transmission-capacity communication. The system integrates a multi-mode fiber (MMF) cable with free space optics (FSO) using orbital angular momentum (OAM) multiplexing. The proposed system uses a single wavelength to simultaneously transmit 40 Gb/s data on four distinct OAM beams ((LG0,0, LG0,17, LG0,40, and LG0,75)) with a symmetric circular shape. The performance of the system was evaluated with a fixed MMF length of 100 m and varying FSO ranges, according to extreme climate changes such as rain and fog, and atmospheric turbulences. The results indicate that the suggested system is capable of transmit up to 1250 m (100 m MMF length + 1150 m FSO range) under clear weather (CW) and weak turbulence (WT), with a minimum transmission distance of 440 m (100 m MMF length + 340 m FSO range) under heavy fog (HF). The performance of the system was also demonstrated based on actual meteorological data for two different cities, Alexandria, Egypt and Chandigarh, India, with different visibility ranges and geographical locations. Since Alexandria has a longer visibility range than Chandigarh, our proposed model could spread over a larger area than it could be in Chandigarh. The obtained ranges were within an acceptable bit error rate (BER) of less than 10−9. The system’s maximum range was 1200 m for Alexandria and 1100 m for Chandigarh. The proposed model can be suggested for use in next generation (NG) passive optical networks (PON) for supporting symmetric 40 Gb/s data.