Understanding the brittle fracture behavior of rock is crucial for engineering and Earth science. In this paper, based on acoustic emission (AE) and laser Doppler vibration (LDV) monitoring technology, the staged damage behaviors of rock-like materials with different brittleness degrees under uniaxial compression are studied via multiple parameters. The results show that the brittleness degree determines the fracture mode. As the specimen’s brittleness degree increases, the tensile failure increases and shear failure decreases. AE activity is enhanced at the crack damage point. With an increasing specimen brittleness degree, different instability precursor information is shown during the unstable crack growth stage: the AE b value changes from the fluctuating to continuously decreasing state, and the natural frequency changes from the stable fluctuation to upward fluctuation state. The AE b value near the stress drop is the smallest, and it decreases with an increasing brittleness degree. The natural frequency reduction indicates the rock-like fracture. The natural frequency is a symbolic index that reflects staged damage characteristics and predicts the amount of energy released by brittle failure. These findings provide guidelines for rock stability monitoring and provide support for better responses to stability evaluations of rock slopes, rock collapses, and tunnel surrounding rock in engineering.

Landslides are frequent and destructive geological disasters. Accurate landslide identification is essential for post-disaster reconstruction and preventing secondary disasters. Deep learning has shown considerable potential for recognizing landslide objects from remote-sensing images; however, existing models still suffer from insufficient detection accuracy in scenarios with complex backgrounds, blurred boundary localization, sample-class imbalance, and difficulty in balancing detection speed and segmentation accuracy. To address these issues, this study investigates landslide identification in the Great Bend of the Yarlung Zangbo River region using improved deep learning models and heterogeneous optical remote-sensing imagery. (1) By introducing the convolutional block attention module (CBAM) into YOLOv8, 82.79% precision was achieved, and the recall improved by 9.56% compared to the original model, reaching a mean average precision of 75.27% while maintaining computational efficiency, outperforming YOLOv5 and the original YOLOv8. (2) Replacing the cross-entropy loss with Focal Loss in DeepLabV3+ improved the landslide edge segmentation by dynamically adjusting the weights of difficult and easy samples. Compared with the original DeepLabV3+, the precision and recall of the DeepLabV3+-FL semantic-segmentation model were improved by 0.26% and 1.93%, respectively, with the mean pixel accuracy and mean intersection over union reaching 81.76% and 62.59%, respectively. Overall, the two improved models enhanced the accuracy of landslide identification and resistance to interference, demonstrating potential for landslide monitoring and emergency response.