Zhuoran Wang

Frequency-Modulated Continuous-Wave (FMCW) Laser Detection and Ranging (LiDAR) systems are widely used due to their high accuracy and resolution. Nevertheless, conventional distance extraction methods often lack robustness in noisy and complex environments. To address this limitation, we propose a deep learning-based signal extraction framework that integrates a Dual Convolutional Neural Network (DCNN) with a Transformer model. The DCNN extracts multi-scale spatial features through multi-layer and pointwise convolutions, while the Transformer employs a self-attention mechanism to capture global temporal dependencies of the beat-frequency signals. The proposed DCNN–Transformer network is evaluated through beat-frequency signal inversion experiments across distances ranging from 3 m to 40 m. The experimental results show that the method achieves a mean absolute error (MAE) of 4.1 mm and a root-mean-square error (RMSE) of 3.08 mm. These results demonstrate that the proposed approach provides stable and accurate predictions, with strong generalization ability and robustness for FMCW LiDAR systems.

Frequency-Modulated Continuous Wave (FMCW) Laser Detection and Ranging (LiDAR) systems have been widely utilized for their high accuracy and high resolution. However, traditional distance extraction methods often suffer from insufficient robustness in high-noise and complex background environments. To address this issue, this manuscript proposes a deep learning-based signal information extraction method that integrates a Dual Convolutional Neural Network (DCNN) with a Transformer model to enhance system performance. The DCNN employs multi-layer deep convolutions and pointwise convolutions to meticulously extract multi-scale spatial features, while the Transformer leverages the self-attention mechanism to efficiently capture the global temporal dependencies of the beat frequency signals. The proposed DCNN-Transformer network is applied to beat frequency signal inversion experiments at various distances. Experimental results on the test dataset covering a ranging distance from 3m to 40m demonstrate that the proposed method achieves a mean absolute error (MAE) of 4.1mm and a root mean square error (RMSE) of merely 3.08mm. The results indicate that our method is capable of achieving excellent prediction stability and high accuracy, exhibiting strong generalization capability and robustness for FMCW Lidar system.

The frequency linearity of a frequency-swept light signal is critical for ensuring the precision of Frequency-Modulated Continuous-Wave (FMCW) laser ranging systems. A two-stage nonlinearity correction mechanism for frequency-swept light is proposed, combining both data-driven and principle-based approaches. In the main correction stage utilizing an electro-optic phase-locked loop (EO-PLL), high temporal resolution phase detection is achieved. To address the failure of the EO-PLL caused by a bandwidth limitation of the digital loop filter (DLF), a novel pre-correction mechanism is developed based on a data-driven approach. In this mechanism, the neural network (NN) model establishes a mapping relationship between the input and output of the real laser-modulation system, which effectively simulates this physical system and avoids the risk of trial-and-error damage. Afterwards, the Soft Actor–Critic (SAC) model interacts with the NN model and trains a decision-making agent to determine the optimal modulation strategy for the nonlinearity pre-correction of the frequency-swept light. During the training process of the SAC agent, both the modulation strategy and the accuracy of evaluating the strategy’s effectiveness are optimized. Moreover, in contrast to the basic Actor–Critic model, the SAC model enhances the exploration of modulation possibilities by maximizing entropy expectation of random strategy, thereby improving the robustness of the pre-correction mechanism. Finally, the frequency-swept characteristic analysis experiment proves that integrating NN-SAC with EO-PLL enables frequency locking under the reduced bandwidth of the DLF. Additionally, through actual ranging experiments, it is also demonstrated that the proposed mechanism significantly enhances ranging precision, repeatability, and stability. Therefore, by integrating data-driven and principle-based approaches, this investigation offers an innovative perspective for the nonlinearity correction of FMCW laser ranging and, furthermore, electro-optic control scenarios.

A precise conveyor belt deviation monitoring method using line array point cloud data is proposed and demonstrated, which can ensure the healthy running of the conveyor system. The point cloud data characterizing the surface of the conveyor belt is collected in a line scanning way. Then, using a unique soft extraction method that weighted fusing three key features (cross-sectional variation, belt’s horizontal width, and previous frame) to process this data, the edge information of the conveyor belt can be accurately and robustly identified in real-time. Furthermore, the point cloud processing mode enables a belt-segmented deviation analysis method based on a standard sequence query. This can accurately determine the offset value and deviation trend of the conveyor belt, thereby achieving early warning of deviation faults. Experimental results show that the belt edge identification precision can reach 0.3 mm, and an early warning can be provided at least 57 m before the occurrence of a belt deviation fault. This belt deviation monitoring method can be widely applied in various working environments, especially in harsh conditions like mines and ports. It also has potential applications in automated production lines within Industry 4.0.

Many major diseases can cause changes in the morphology of blood vessels, and the segmentation of retinal blood vessels is of great significance for preventing these diseases. Obtaining complete, continuous, and high-resolution segmentation results is very challenging due to the diverse structures of retinal tissues, the complex spatial structures of blood vessels, and the presence of many small ships. In recent years, deep learning networks like UNet have been widely used in medical image processing. However, the continuous down-sampling operations in UNet can result in the loss of a significant amount of information. Although skip connections between the encoder and decoder can help address this issue, the encoder features still contain a large amount of irrelevant information that cannot be efficiently utilized by the decoder. To alleviate the irrelevant information, this paper proposes a feature selection module between the decoder and encoder that utilizes the self-attention mechanism of transformers to accurately and efficiently select the relevant encoder features for the decoder. Additionally, a lightweight Residual Global Context module is proposed to obtain dense global contextual information and establish dependencies between pixels, which can effectively preserve vascular details and segment small vessels accurately and continuously. Experimental results on three publicly available color fundus image datasets (DRIVE, CHASE, and STARE) demonstrate that the proposed algorithm outperforms existing methods in terms of both performance metrics and visual quality.