Clouds are one of the most important meteorological phenomena affecting the Earth radiation balance. The increasing development of whole-sky images enables temporal and spatial high-resolution sky observations and provides the possibility to understand and quantify cloud effects more accurately. In this letter, an attempt has been made to examine the machine learning [multilayer perceptron (MLP) neural networks and support vector machine (SVM)] capabilities for automatic cloud detection in whole-sky images. The approaches have been tested on a significant number of whole-sky images (containing a variety of cloud overages in different seasons and at different daytimes) from Vigna di Valle and Tor Vergata test sites, located near Rome. The pixel values of red, green, and blue bands of the images have been used as inputs of the mentioned models, while the outputs provided classified pixels in terms of cloud coverage or others (cloud-free pixels and sun). For the test data set, the overall accuracies of 95.07%, with a standard deviation of 3.37, and 93.66%, with a standard deviation of 4.45, have been obtained from MLP neural networks and SVM models, respectively. Although the two approaches generally generate similar accuracies, the MLP neural networks gave a better performance in some specific cases where the SVM generates poor accuracy.

A multilayer perceptron neural network cloud mask for Meteosat Second Generation SEVIRI (Spinning Enhanced Visible and Infrared Imager) images is introduced and evaluated. The model is trained for cloud detection on MSG SEVIRI daytime data. It consists of a multi-layer perceptron with one hidden sigmoid layer, trained with the error back-propagation algorithm. The model is fed by six bands of MSG data (0.6, 0.8, 1.6, 3.9, 6.2 and 10.8 μm) with 10 hidden nodes. The multiple-layer perceptrons lead to a cloud detection accuracy of 88.96%, when trained to map two predefined values that classify cloud and clear sky. The network was further evaluated using sixty MSG images taken at different dates. The network detected not only bright thick clouds but also thin or less bright clouds. The analysis demonstrated the feasibility of using machine learning models of cloud detection in MSG SEVIRI imagery.

Accurate spatial information of agricultural fields is important for providing actionable information to farmers, managers, and policymakers. On the other hand, the automated detection of field boundaries is a challenging task due to their small size, irregular shape and the use of mixed-cropping systems making field boundaries vaguely defined. In this paper, we propose a strategy for field boundary detection based on the fully convolutional network architecture called ResU-Net. The benefits of this model are two-fold: first, residual units ease training of deep networks. Second, rich skip connections within the network could facilitate information propagation, allowing us to design networks with fewer parameters but better performance in comparison with the traditional U-Net model. An extensive experimental analysis is performed over the whole of Denmark using Sentinel-2 images and comparing several U-Net and ResU-Net field boundary detection algorithms. The presented results show that the ResU-Net model has a better performance with an average F₁ score of 0.90 and average Jaccard coefficient of 0.80 in comparison to the U-Net model with an average F₁ score of 0.88 and an average Jaccard coefficient of 0.77.

Nowadays, the unmanned aerial vehicle (UAV) has a wide application in transportation. For instance, by leveraging it, we are able to perform accurate and real-time vehicle speed detection in an IoT-based smart city. Although numerous vehicle speed estimation methods exist, most of them lack real-time detection in different situations and scenarios. To fill the gap, this paper introduces a novel low-altitude vehicle speed detector system using UAVs for remote sensing applications of smart cities, forging to increase traffic safety and security. To this aim, (1) we have found the best possible Raspberry PI’s field of view (FOV) camera in indoor and outdoor scenarios by changing its height and degree. Then, (2) Mobile Net-SSD deep learning model parameters have been embedded in the PI4B processor of a physical car at different speeds. Finally, we implemented it in a real environment at the JXUST university intersection by changing the height (0.7 to 3 m) and the camera angle on the UAV. Specifically, this paper proposed an intelligent speed control system without the presence of real police that has been implemented on the edge node with the configuration of a PI4B and an Intel Neural Computing 2, along with the PI camera, which is armed with a Mobile Net-SSD deep learning model for the smart detection of vehicles and their speeds. The main purpose of this article is to propose the use of drones as a tool to detect the speeds of vehicles, especially in areas where it is not easy to access or install a fixed camera, in the context of future smart city traffic management and control. The experimental results have proven the superior performance of the proposed low-altitude UAV system rather than current studies for detecting and estimating the vehicles’ speeds in highly dynamic situations and different speeds. As the results showed, our solution is highly effective on crowded roads, such as junctions near schools, hospitals, and with unsteady vehicles from the speed level point of view.

Grassland contributes to carbon storage and animal feed production. Its yield is largely determined by the cutting times of grassland. Previous studies have used remote sensing data for grassland biomass estimation, but only a few studies have focused on SAR remote sensing approaches for automatic grassland cutting status detection. Due to the occurrence of multiple cuttings in a year, it is crucial to effectively monitor grassland cutting events in order to achieve accurate biomass estimations of a whole season. In this study, we examined the capabilities of multilayer perceptron neural networks for automatic grassland cutting status detection using SAR imagery. The proposed model inputs are a time series dataset of VV and VH Sentinel-1 C-band SAR and second-order texture metrics (homogeneity, entropy, contrast and dissimilarity). The proposed approach has been successfully tested on a dataset collected from several fields in Germany in 2016, with an overall accuracy of 85.71% for the validation set.

Dark-spot detection is a critical step in oil-spill detection. In this paper, a novel approach for automated dark-spot detection using synthetic aperture radar imagery is presented. A new approach from the combination of Weibull multiplicative model (WMM) and pulse-coupled neural network (PCNN) techniques is proposed to differentiate between the dark spots and the background. First, the filter created based on WMM is applied to each subimage. Second, the subimage is segmented by PCNN techniques. As the last step, a very simple filtering process is used to eliminate the false targets. The proposed approach was tested on 60 Envisat and ERS2 images which contained dark spots. The same parameters were used in all tests. For the overall data set, an average accuracy of 93.66% was obtained. The average computational time for dark-spot detection with a 512 × 512 image is about 7 s using IDL software, which is the fastest one in this field at present. Our experimental results demonstrate that the proposed approach is very fast, robust, and effective. The proposed approach can be applied on any kind of synthetic aperture radar imagery.