In solving change detection problem, unsupervised methods are usually preferred to their supervised counterparts due to the difficulty of producing labelled data. Nevertheless, in this paper, a supervised deep learning‐based method is presented for change detection in synthetic aperture radar (SAR) images. A Deep Belief Network (DBN) was employed as the deep architecture in the proposed method, and the training process of this network included unsupervised feature learning followed by supervised network fine‐tuning. From a general perspective, the trained DBN produces a change detection map as the output. Studies on DBNs demonstrate that they do not produce ideal output without a proper dataset for training. Therefore, the proposed method in this study provided a dataset with an appropriate data volume and diversity for training the DBN using the input images and those obtained from applying the morphological operators on them. The great computational volume and the time‐consuming nature of simulation are the drawbacks of deep learning‐based algorithms. To overcome such disadvantages, a method was introduced to greatly reduce computations without compromising the performance of the trained DBN. Experimental results indicated that the proposed method had an acceptable implementation time in addition to its desirable performance and high accuracy.

This study proposed a novel algorithm for coastline detection in single-polarization SAR images based on local spectral histogram (LSH) and the level set method (LSM). The proposed algorithm includes two main steps. In the processing step, a rough land/sea segmentation is done by utilizing a texturebased segmentation using LSH. In the post-processing step, the region-based LSM is used to refine the previous segmentation and extract the coastline accurately. The level set (LS) is initialized using the results of LSH segmentation. This prevents the development of spurious contours, becoming trapped in the local minimum, eliminates the manual support and also speeds up the LS evolution. A hierarchical LS regularization is proposed using two Gaussian kernels which is compatible with noisy images and able to detect very narrow regions. The proposed algorithm is able to detect a coastline at the full-resolution of the input SAR image and also is robust to noise. It does not require pre-processing step to reduce speckle. A criterion to quantify the accuracy of the results was also proposed. The experimental results for a number of real high-resolution single-polarization SAR images demonstrate that the proposed method is robust to noise and efficient for coastline detection in different coastal and sea environments.

Coastline extraction from synthetic aperture radar (SAR) data is difficult because of the presence of speckle noise and strong signal returns from the wind-roughened and wave-modulated sea surface. High resolution and weather change independent of SAR data lead to better monitoring of coastal sea. Therefore, SAR coastline extraction has taken up much interest. The active contour method is an efficient algorithm for the edge detection task; however, applying this method to high-resolution images is time-consuming. The current article presents an efficient approach to extracting coastlines from high-resolution SAR images. First, fuzzy clustering with spatial constraints is applied to the input SAR image. This clustering method is robust for noise and shows good performance with noisy images. Next, binarization is carried out using Otsu’s method on the fuzzification results. Third, morphological filters are used on the binary image to eliminate spurious segments after binarization. To extract the coastline, an active contour level set method is used on the initial contours and is applied to the input SAR image to refine the segmentation. Because the proposed approach is based on an active contour model, it does not require preprocessing for SAR speckle reduction. Another advantage of the proposed method is the ability to extract the coastline at full resolution of the input SAR image without degrading the resolution. The proposed approach does not require manual initialization for the level set method and the proposed initialization speeds up the level set evolution. Experimental results on low- and high-resolution SAR images showed good performance for coastline extraction. A criterion based on neighbourhood pixels for the coastline is proposed for the quantitative expression of the accuracy of the method.

As one of the non-contact methods for incipient fault diagnosis of power distribution lines, UV-Visible imaging has become popular due to its good performance and robustness against environmental parameters. This paper presents a deep learning-based method for UV-Visible video processing. First, for preparing the dataset, some videos are acquired from distribution lines using CoroCam 6D2 considering observation distance, the imager's gain, air pressure, and humidity as effective parameters on the discharge area in the images. Then, based on the logged information of unplanned power outages in the system and also planned inspections over a year, the incipient fault type and the severity level of the incipient fault are determined for each video. For processing each video, frames are extracted with the rate according to the nominal voltage of the line. Power devices are detected in each frame using Faster R-CNN and are tracked through the whole video frames to compensate the camera movement. Then, color thresholding is used for each device to identify corona discharges in the frames. In-depth median filtering is also used through the video to eliminate noises in the UV channel. Finally, based on the ratio of the detected discharge area to the area of the equipment, the severity level of the incipient fault is determined. The proposed method not only performs better than the state-of-the-art but also is a practical method, and with the least dependence on environmental conditions, can automatically identify incipient faults in distribution lines, even in videos containing several possible defective devices.

This paper presents a deep learning-based method for defect detection and classification of power distribution lines using video analysis. The first stage is dataset preparation in which different videos are recorded by a CoroCam 6D2 camera from distribution lines and labeled based on the defect type and the severity level of the defect. Then, a process is followed in such a way that a limited number of frames are used for processing, and power devices are detected in each frame using Faster R-CNN. The detector performance is improved by increasing the training dataset and changing the training hyper-parameters. Next, an equipment tracking technique is applied through the video frames. In the following, a method based on the classification of power equipment is presented. In this method, by corona color thresholding in all frames and applying a median filter over time, the connected components representing the corona are identified. Then, the frame containing the closest component to the component in the median image is selected. In the selected frame, the area around the component is cut and given to AlexNet or GoogleNet to determine the equipment type and the severity level of defect is determined. Also, the defective equipment phase is determined based on the insulators detected in the video. The proposed method not only performs better than the state-of-the-art but also is a practical method, and with the least dependence on environmental conditions, can automatically identify defects in distribution lines, even in videos containing several possible defective devices.