Point cloud registration serves as a critical tool for constructing 3D environmental maps. Both geometric and color information are instrumental in differentiating diverse point features. Specifically, when points appear similar based solely on geometric features, rendering them challenging to distinguish, the color information embedded in the point cloud carries significantly important features. In this study, the colored point cloud is utilized in the FCGCF algorithm, a refined version of the FCGF algorithm, incorporating color information. Moreover, we introduce the PointDSCC method, which amalgamates color consistency from the PointDSC method for outlier removal, thus enhancing registration performance when synergized with other pipeline stages. Comprehensive experiments across diverse datasets reveal that the integration of color information into the registration pipeline markedly surpasses the majority of existing methodologies and demonstrates robust generalizability.

Single Image Defocus Deblurring (SIDD) remains challenging due to spatially varying blur kernels, particularly in processing high-resolution images where traditional methods often struggle with artifact generation, detail preservation, and computational efficiency. This paper presents Swin-Diff, a novel architecture integrating diffusion models with Transformer-based networks for robust defocus deblurring. Our approach employs a two-stage training strategy where a diffusion model generates prior information in a compact latent space, which is then hierarchically fused with intermediate features to guide the regression model. The architecture incorporates a dual-dimensional self-attention mechanism operating across channel and spatial domains, enhancing long-range modeling capabilities while maintaining linear computational complexity. Extensive experiments on three public datasets (DPDD, RealDOF, and RTF) demonstrate Swin-Diff’s superior performance, achieving average improvements of 1.37% in PSNR, 3.6% in SSIM, 2.3% in MAE, and 25.2% in LPIPS metrics compared to state-of-the-art methods. Our results validate the effectiveness of combining diffusion models with hierarchical attention mechanisms for high-quality defocus blur removal.

In the context of real-world Internet of Things (IoT) implementations, the delivery of images within blockchain environments is becoming increasingly crucial. However, the multitable hashing technologies currently used for retrieval suffer from considerable redundancy in tables, leading to diminished performance of the distributed system. To address these challenges, we propose a hash code selection-based secure image retrieval (HCSSIR) scheme for data sharing and secure retrieval on the blockchain. First, we introduce a learning-based method to shorten hash codes. Initial long hash codes are generated using locality-sensitive hashing families for high efficiency and accuracy. The most informative hash tables are then selected based on a learning model that has been pretrained on a dataset specifically prepared for this purpose. This approach aims to reduce space and computational requirements, potentially further decreasing time consumption. Next, a cryptographic hierarchical index structure based on the compressed hash codes is designed, along with its construction and searching methods. We also create a secure Hamming distance computation algorithm to preserve the index features and access patterns. Finally, proxy re-encryption is adapted to ensure data availability in multiuser settings. We offer a security analysis to confirm that the HCSSIR scheme effectively protects the privacy of images, indexes, and queries. Furthermore, the proposed selection method achieves an approximate 12.3% enhancement in F1-score for code compression when compared with the cutting-edge method. Additionally, the HCSSIR scheme registers improvements of approximately 27.58% and 56.7% in searching and storage efficiencies, respectively. These results underscore its suitability for real-world IoT applications.