Zhongyuan Lu

Traditional object detection models often lose the detailed outline information of the object. To address this problem, we propose the Fourier Series Object Detection (FSD). It encodes the object's outline closed curve into two one-dimensional periodic Fourier series. The Fourier Series Model (FSM) is constructed to regress the Fourier series for each object in the image. Thus, during inference, the detailed outline information of each object can be retrieved. We introduce Rolling Optimization Matching for Fourier loss to ensure that the model's learning process is not affected by the sequence of the starting points of the labeled contour points, speeding up the training process. The FSM demonstrates improved feature extraction and descriptive capabilities for non-rectangular or elongated object regions. The model achieves AP50 = 73.3% on the DOTA 1.5 dataset, which surpasses the state-of-the-art (SOTA) method by 6.44% at 66.86%. On the UCAS dataset, the model achieves AP50 = 97.25%, also surpassing the performance indicators of the SOTA methods. Furthermore, we introduce the object's Fourier power spectrum to describe outline features and the Fourier vector to indicate its direction. This enhances the scene semantic representation of the object detection model and paves a new pathway for the evolution of object detection methodologies.

PointNet, the first deep learning framework for processing 3D point clouds, suffers from significant information loss during point cloud classification due to its Max pooling method, leading to decreased performance and accuracy when handling large-scale point clouds. In this paper, we propose an improved method for PointNet classification accuracy—a Functional Multi-head Pooling encoding model. Based on the permutation-invariant functional concept, this model constructs richer multi-head invariant feature vectors, significantly enhancing recognition classification scores, and introducing a new model—Functional Multi-head Pooling (FMP)—capable of reducing overfitting. Due to the non-monotonicity, continuous differentiability, and controllable amplitude of functionals, we conducted experiments using cosine and sine trigonometric functions to replace the baseline Max method. The new approach also involves aggregating the results of the cosine and sine transformations, which allows for retaining more features during training compared to PointNet, thereby enhancing the training model’s capabilities. Our three rounds of experiments on two datasets including ModelNet40 have proven the effectiveness of the functional approach. The new method can increase accuracy by up to 3.25% on the same datasets and significantly improving generalization capabilities.

In hyperspectral imaging (HSI), stripe noise is one of the most common noise types that adversely affects its application. Convolutional neural networks (CNNs) have contributed to state-of-the-art performance in HSI destriping given their powerful feature extraction and learning capabilities. However, it is difficult to obtain paired training samples for real data. Most CNN destriping methods construct a paired training dataset with simulated stripe noise for network training. However, when the stripe noise of real data is complex, destriping performance of the model is constrained. To solve this problem, this study proposes a real HSI stripe removal method using a toward real HSI stripe removal via direction constraint hierarchical feature cascade network (TRS-DCHC). TRS-DCHC uses the stripe noise extract subnetwork to extract stripe patterns from real stripe-containing HSI data and incorporates clean images to form paired training samples. The destriping subnetwork advantageously utilizes a wavelet transform to explicitly decompose stripe and stripe-free components. It also adopts multi-scale feature dense connections and feature fusion to enrich feature information and deeply mine the discriminate features of stripe and stripe-free components. Our experiments on both simulated and real data of various loads showed that TRS-DCHC features better performance in both simulated and real data compared with state-of-the-art method.