Obtaining real-time, objective, and high-precision distribution information of surface cracks in mining areas is the first task for studying the development regularity of surface cracks and evaluating the risk. The complex geological environment in the mining area leads to low accuracy and efficiency of the existing extracting cracks methods from unmanned air vehicle (UAV) images. Therefore, this manuscript proposes a new identification method of surface cracks from UAV images based on machine learning in coal mining areas. First, the acquired UAV image is cut into small sub-images, and divided into four datasets according to the characteristics of background information: Bright Ground, Dark Dround, Withered Vegetation, and Green Vegetation. Then, for each dataset, a training sample is established with cracks and no cracks as labels and the RGB (red, green, and blue) three-band value of the sub-image as feature. Finally, the best machine learning algorithms, dimensionality reduction methods and image processing techniques are obtained through comparative analysis. The results show that using the V-SVM (Support vector machine with V as penalty function) machine learning algorithm, principal component analysis (PCA) to reduce the full features to 95% of the original variance, and image color enhancement by Laplace sharpening, the overall accuracy could reach 88.99%. This proves that the method proposed in this manuscript can achieve high-precision crack extraction from UAV image.

Widespread ground fissures caused by coal mining subsidence are a main cause of ecological destruction in coal mining areas, and the rapid monitoring of ground fissures is essential for ecological restoration. Traditional fissure monitoring technologies are time consuming and laborious. Therefore, we developed a method to automatically extract ground fissures from high-resolution UAV images. First, a multiscale Hessian-based enhancement filter was utilized to enhance the ground fissures in grayscale images. Then, a simple single-thresholding operation was applied to segment the enhanced image to generate a binary ground fissure map. Finally, incomplete path opening was performed to eliminate the noises in the fissure extraction results. We selected the N1212 working face of the Ningtiaota Coal Mine in Shenmu County, China, as the study area. The results indicated that the ranges of correctness, completeness, and the kappa coefficient of the extracted results were 66.23–79.00%, 69.03–73.22%, and 67.91–75.88%, respectively. Image resolution is the key factor for successful fissure detection; the method proposed in this paper can extract ground fissures with a width greater than one pixel (2.64 cm), and the detection ratio for fissures with a width greater than two pixels was over 87%. Our research has solved the problem of the rapid monitoring of ground fissures to a certain extent and can act as a valuable tool for ecological restoration in mining areas.

In order to effectively control the damage caused by surface cracks to a geological environment, we need to find a convenient, efficient, and accurate method to obtain crack information. The existing crack extraction methods based on unmanned air vehicle (UAV) images inevitably have some erroneous pixels because of the complexity of background information. At the same time, there are few researches on crack feature information. In view of this, this article proposes a surface crack extraction method based on machine learning of UAV images, the data preprocessing steps, and the content and calculation methods for crack feature information: length, width, direction, location, fractal dimension, number, crack rate, and dispersion rate. The results show that the method in this article can effectively avoid the interference by vegetation and soil crust. By introducing the concept of dispersion rate, the method combining crack rate and dispersion rate can describe the distribution characteristics of regional cracks more clearly. Compared to field survey data, the calculation result of the crack feature information in this article is close to the true value, which proves that this is a reliable method for obtaining quantitative crack feature information.

Underground coal mining inevitably causes ground fissures, especially permanent cracks that cannot be closed at the boundary of the working face. Studying the underground three-dimensional morphology of the permanent cracks allows one to accurately constrain the formation and development of the ground fissures. This information will contribute to reducing mine disasters and is also a prerequisites to avoid environmental pollution. We selected the Zhangjiamao coal mine (China), which is situated in a collapsible loess area, as a case study for deciphering the formation of permanent cracks. After injecting gypsum slurry into the mine, a three-dimensional model of the ground fissures is obtained by three-dimensional (3D) laser scanner technology that records the 3D underground morphology. Integrating the geological context of a collapsible loess area, the characteristics and main processes of the ground fissure development are constrained: (1) The width of the ground fissure decreases to 0 with increasing depth and is strongly affected by the soil composition. (2) Along the vertical extension direction, the ground fissures are generally inclined to the inner-side of the working face, but the direction remains uncertain at different depths. (3) The transverse propagation direction of the ground fissure becomes more complex with increasing depth. (4) Under the influence of soil texture and water, loose soil fills the bottom of the ground fissure, thus affecting the underground 3D morphology.

Biological soil crust is an important feature of desert ecosystem composition and surface landscape. Determining the role of biological soil crust in the energy flow and logistics cycle in the desert ecosystem is one of the frontier areas of ecological restoration in arid and semi-arid areas. Obtaining biological soil crust information from drone images is an efficient, fast, and low-cost method. However, due to the scattered and uneven growth of biological soil crusts and the complexity of the field environment, it is difficult to accurately extract biological soil crusts. In view of this, this study used the improved UNet++ model to extract biological soil crusts based on UAV image data. Firstly, the optimal Epoch, Backbone, and Loss function are selected and trained based on the network structure of UNet++ model. Then, the improved UNet++ model proposed in this paper, which takes ResNeXt as the Backbone and Soft Cross-Entropy Loss+Dice Loss as the Loss Function, is obtained. Finally, the test results of UNet++, U-Net, LinkNet, FPN, and PSPNet are compared with the improved UNet++ model in this paper. The results showed that the improved UNet++ model had the best segmentation effect, and the precision, recall, F1-Score, and IoU were 0.9788, 0.9501, 0.9495, and 0.9309, respectively. UAV image biological soil crust recognition based on the improved UNet++ model in this paper can obtain high-precision extraction results, provide good data support for studying the development of biological soil crusts in arid areas, and provide a new method for precise segmentation of different features in arid areas. In particular, it is of great significance to evaluate the governance effect of ecological restoration projects.

Obtaining real-time, objective, and high-precision distribution information of surface cracks in mining areas is the first task for studying the development regularity of surface cracks and evaluating their risk. However, the complex surface environment causes low precision of crack extraction in the existing unmanned aerial vehicle (UAV) image methods. Therefore, we propose an extraction method for surface cracks from UAV sub-images after cutting and a new evaluation index of APS (Accuracy-Precision-Sensitivity). First, the UAV image was divided into sub-images with unit sizes of 50, 100, and 200, and the bare ground (BG) and vegetation (VT) datasets based on background characteristics were built. Then, we used the five crack extraction methods improved and proposed to process the datasets. Finally, this study determined the optimal unit size to extract the cracks and the optimal method for a certain size by comparison. The results showed that the extraction method based on the unit method was better than the complete UAV image, and the crack extraction effect of the BG dataset was significantly better than that of the VT dataset. The APS index, which comprehensively considers the accuracy, precision, and sensitivity, is obviously more reasonable than the accuracy index for evaluating crack extraction methods. Moreover, the smaller the unit size, the better the effect of crack extraction. The Hue-Saturation-Value threshold segmentation method had the best extraction effect when the size was small, and the deep learning method was the best when the size was large. We proved that the proposed method can achieve higher-precision crack extraction from UAV images and can also support data calculation in crack feature extraction.