The method of detecting ballast bed defects using ground penetrating radar (GPR) is an important method for guiding the maintenance of railway infrastructure. Currently, this technology primarily relies on time–frequency analysis to assess the condition of the ballast bed and manual interpretation of GPR images to identify defect areas and types, resulting in low automation levels. This paper proposes a bimodal deep learning classification model that enables intelligent classification of moisture and mud pumping defects in ballast beds. This model includes two channels, each processing a different data modality. One channel uses a Multilayer Perceptron (MLP) to extract features of A-scan data in the time domain. The other channel utilizes Short-Time Fourier Transform (STFT) to convert time domain signals into frequency domain signals, which are then processed by a ResNet18 to extract frequency domain features. By fusing the time and frequency features, the proposed Time-Frequency-Fusion ResNet model (TFF-ResNet) demonstrates superior performance. Experimental results show that TFF-ResNet outperforms the standalone MLP and ResNet18 models, with performance improvements of approximately 24% and 14% on the validation dataset, and 21% and 34% on the testing dataset, respectively.

In the field of solid-state luminescence, Cu²⁺ has long been widely acknowledged for its capacity to emit infrared light. However, the occurrence of visible emission from Cu²⁺ ions had been infrequently observed and reported. In this study, we made an intriguing discovery by examining the behavior of Cu²⁺ within an irregular coordination environment of Ba in BaGa₂O₄. When excited by UV light, Cu²⁺ unexpectedly gave a vibrant yellow–red emission, covering a wavelength range spanning from 500 to 750 nm. More noteworthy, by simply manipulating the excitation wavelength or adjusting the temperature, the peak wavelength of the emission could be effectively tuned from approximately 600 to 660 nm, which could be attributed to the luminescence nature of the charge transfer (CT) between O²⁻ and Cu²⁺. Moreover, the phosphor material displayed a remarkable persistent luminescence (PerL) lasting up to 12 h after UV light excitation. Through thermoluminescence (TL) measurements and first-principle calculations, we found that the intrinsic defects, such as vacancies of oxygen and gallium (V_O and V_Ga^″), played important roles for the PerL phenomena. These findings highlighted the exceptional tunability and PerL properties of BaGa₂O₄:Cu²⁺. Our study provided a new potential guideline for the design of Cu²⁺-activated phosphors in visible region, and opened up new avenues for the research in related functional luminescence materials.

The red afterglow of current rare-earth-activated long persistent luminescence (LPL) phosphors is largely still less than 6 h, in contrast to the 20 or 30 h long blue- or green-emitting ones, becoming the main obstacle to realize their multiscenario applications in practice. Herein, we report a rare-earth-free red-emitting LPL phosphor SrGa₂O₄:Cu²⁺that can exhibit an afterglow at about 622 nm lasting over 30 h, which can largely match with the luminance of current blue- or green-emitting LPL phosphors. We find that the Cu²⁺ion could be charged by ultraviolet light from 280 to 420 nm, and the emission has a very broad band with a full width at half-maximum of about 150 nm. Combining the thermoluminescence measurement and the first-principles calculation, we find that the O vacancies and the -1 charged Ga vacancies could store the holes and contribute to the LPL of SrGa₂O₄:Cu²⁺. Our results may dramatically promote and expand its potential applications and stimulate the research of the multicolor LPL phosphors in future.