Unexploded ordnance (UXO) presents a significant risk to both the natural environment and human safety. Current deep learning detection mechanisms are characterized by limited interpretability, resulting in a persistent absence of detection methods that are simultaneously efficient, secure, and precise. To address these challenges, a dual-mode detection method based on interpretation-promoted YOLOv5 is proposed in this paper. A comprehensive dataset has been constructed to focus on the scarcity of low-altitude UXO target datasets, incorporating both visible light and infrared imagery. Dataset augmentation has been implemented through the application of generative adversarial networks for image super-resolution reconstruction, thereby enhancing the robustness of the dataset. To tackle issues of low detection accuracy and inadequate background discrimination associated with a single information source, an integration of visible light and infrared data has been proposed to enhance the interpretability of the YOLOv5 algorithm, leading to improved detection performance. Extensive low-altitude detection experiments were conducted in field environments using UAVs. Experimental results demonstrate that the proposed method achieves a remarkable detection accuracy of up to 97.1% and an impressive detection speed of up to 60.3 frames per second.

To address the issue of weak geoelectric current field signals that are severely affected by noise and cannot be directly used for geological structure analysis in injected geoelectric current field detection technology, this study proposes a complete ensemble empirical mode decomposition with adaptive noise and improved wavelet thresholding collaborative denoising (CEEMDAN-IWT) method to enhance the interpretation accuracy of geoelectric current signals. The method performs signal decomposition through CEEMDAN and selects the effective intrinsic mode function (IMF) components based on the variance contribution criterion for preliminary denoising. It then combines the improved wavelet thresholding function for further fine denoising and reconstruction, obtaining high signal-to-noise ratio (SNR) electrical data. Simulation and real-world data validation show that in a simulation experiment with an initial SNR of −5 dB, the method improves the $SNR$ to 18.65 dB, and the SNR enhancement is superior to traditional methods under various noise intensities. In practical applications, the normalized cross-correlation (NCC) between the denoised signal and the original injected signal reaches as high as 0.9254, significantly outperforming traditional methods. At the same time, it balances the preservation of signal features with noise suppression, offering significant application value for improving the reliability of injected geoelectric current field detection data.

Aiming at the problem of a lack of positioning satellites and no available beacons in underground space, an injected ground electrode current field signal localization method is proposed. An extremely low-frequency current field signal is applied to two pairs of electrodes inserted into the earth to form a ground current field underground, and the ground electrode current field signal detected at the detection end is used for localization, which can effectively provide reference localization for the underground space when the satellite positioning fails. On this basis, considering that the ground electrode current field signal is susceptible to the influence of the geological structure, electromagnetic interference, and the complexity of the propagation path during underground transmission, which results in the signal showing strong non-stationary characteristics, it is difficult for the traditional time–frequency analysis method to accurately extract stable and reliable positioning characteristics. In order to improve the signal-processing accuracy and robustness, this paper introduces fractional Fourier transform (FRFT) to process the detected signals, and focuses the signal energy more effectively under the optimal order. In order to verify the effectiveness of the localization method, several experiments on the localization of ground electrode current field signals are carried out in the underground space. The experimental results show that, in the positioning environment of more than 10,000 square meters, the average positioning error is 6.896 m. The application of this method will provide a solid technical support for life rescue in underground space, provide the ‘last protection’ for rescue, and complete the life chain of emergency first aid, which has an important application prospect and practical value.

Aiming to solve the problems of communication interruption caused by the collapse of underground space, this study constructs a strong penetration information transmission system and proposes a centroid localization method based on the received signal strength indication (RSSI) in an underground space ground electrode current field. This is applicable to localization in underground space such as subways, mines, tunnels, etc., as well as under the environment of collapse. First, the propagation characteristics of the ground current field signal in underground space are analyzed, and the attenuation model of the ground current field signal is constructed by combining the RSSI ranging method. On this basis, an improved weighted centroid localization algorithm is introduced to improve the localization accuracy and reliability by optimizing the algorithm parameters to cope with the fluctuations and instabilities generated in the signal propagation process. The experimental results show that the proposed localization method achieves an average positioning error of 7.47 m in an underground environment of 10,000 square meters, which is 32.32% less compared with the weighted centroid localization algorithm, and 62.74% less compared with the traditional centroid localization algorithm. This method presents a positioning technology that operates independently in underground spaces, overcoming the limitation of traditional wireless positioning systems, which rely on external transmission links. Its application will provide crucial technical support for life-saving operations in underground environments, acting as the ‘last line of defense’ in rescue missions. By completing the emergency response chain, it will enhance disaster rescue capabilities, offering substantial practical value and promising prospects.