In this work, we demonstrated a new type of biomimetic multispectral curved compound eye camera (BM3C) inspired by insect compound eyes for aerial multispectral imaging in a large field of view. The proposed system exhibits a maximum field of view (FOV) of 120 degrees and seven-waveband multispectral images ranging from visible to near-infrared wavelengths. Pinhole imaging theory and the image registration method from feature detection are used to reconstruct the multispectral 3D data cube. An airborne imaging experiment is performed by assembling the BM3C on an unmanned aerial vehicle (UAV). As a result, radiation intensity curves of several objects are successfully obtained, and a land type classification is performed using the K-means method based on the aerial image as well. The developed BM3C is proven to have the capability for large FOV aerial multispectral imaging and shows great potential applications for distant detecting based on aerial imaging.

The compound-eye imaging system emulates the key characteristics of natural compound eyes, including an expansive field of view (FOV) and exceptional sensitivity to moving targets. These inherent properties confer distinct advantages for unmanned reconnaissance applications, facilitating both large-scale monitoring and dynamic object detection tasks. In this work, we present an innovative wide-FOV small object detection method based on the compound-eye imaging system. A convolutional attention super-resolution fusion network (CASFNet) was designed to perform super-resolution upsampling on small target features in images and adaptively fuse multi-layer features, enabling accurate identification of multiple categories of small targets in compound-eye images. In addition, we established what we believe to be a novel compound-eye sub-image (CESI) dataset that utilizes the inherent FOV-overlap among ommatidia to achieve hardware-level data enhancement, providing a robust foundation for model development and validation. Moreover, we introduced a confidence-weighted fusion strategy that exploits system-specific imaging parameters to optimize confidence scores for identical targets across different sub-images. The proposed strategy generates spatially mapped detection results with unified confidence metrics on the reconstructed full-FOV image. Experimental validation demonstrates that the method achieves outstanding performance in multi-category small object detection with a measured precision of 96.2% and mAP of 94.2%, while significantly enhancing the overall reliability of object detection based on the compound-eye imaging system. This advancement paves the way for object detection in wide-area surveillance and intelligent transportation.

Human action recognition aims to understand human behavior and is crucial in enhancing the intelligence and naturalness of human–computer interaction and bionic robots. This paper proposes a method to improve the complexity and real-time performance of action recognition by combining the Kinect sensor with the OpenPose algorithm, the Levenberg–Marquardt (LM) algorithm, and the Dynamic Time Warping (DTW) algorithm. First, the Kinect V2 depth sensor is used to capture color images, depth images, and 3D skeletal point information from the human body. Next, the color image is processed using OpenPose to extract 2D skeletal point information, which is then mapped to the depth image to obtain 3D skeletal point information. Subsequently, the LM algorithm is employed to fuse the 3D skeletal point sequences with the sequences obtained from Kinect, generating stable 3D skeletal point sequences. Finally, the DTW algorithm is utilized to recognize complex movements. Experimental results across various scenes and actions demonstrate that the method is stable and accurate, achieving an average recognition rate of 95.94%. The method effectively addresses issues, such as jitter and self-occlusion, when Kinect collects skeletal points. The robustness and accuracy of the method make it highly suitable for application in robot interaction systems.

Spectro-polarimetric imaging systems can simultaneously acquire spatial, spectral, and polarimetric information during remote sensing, yet the multimodal fusion data are often constrained in practical applications by insufficient exploitation of complementary information across different modalities. To address this issue, we propose a multimodal image fusion method based on a Dual-Branch Cross-Attention Synergistic Transformer (DBCS-T). In our method, three complementary feature components are independently extracted, i.e., a Characteristic Polarization Image (CPI), a Characteristic Spectral Image (CSI) and a Characteristic Intensity Image (CII). For CPI, it is derived from Angle of Linear Polarization (AoLP) and Degree of Linear Polarization (DoLP) inputs via DBCS-T, which integrates a Cross-Channel Transposed Attention (CCTA) module for cross-modal interaction and a Multi-Scale Polarization Feature Adaptive Modulation (MPAM) module for local feature enhancement. CSI is obtained by leveraging maximum-divergence spectral band differences guided by prior spectral radiance curves. CII is computed from the Stokes parameter $S_0$. These three components are then fused via Principal Component Analysis (PCA) into a Multimodal Fusion Image (MFI). To demonstrate the effectiveness of our method, an experiment was conducted on a scene that contains real vegetation, artificial foliage, and same-color metallic objects. The experimental results show that the proposed method achieves effective semantic segmentation of all three target categories. Furthermore, quantitative evaluation demonstrates that the MFI attains the lowest Kullback–Leibler (KL) and Jensen–Shannon (JS) Divergence values among all evaluated modalities, with image entropy exceeding that of individual source inputs. These results validate the complementarity of the extracted multimodal features, significantly enhance the interpretation performance for complex scenes, and demonstrate the broad application potential of the proposed fusion framework in remote sensing and multidimensional imaging.