Non-Destructive Compression Technology for Tool Wear Images in Unmanned Factories

Addressing the real-time monitoring requirements for the wear status of micro-milling tools in the precision manufacturing industry, this paper proposes a lossless compression algorithm based on dual dictionary learning, specifically designed for efficiently compressing images of milling tool wear in unmanned factories. The core of this algorithm lies in utilizing image super-resolution technology to reconstruct lowresolution wear images collected on-site into high-resolution images through an iterative process, thereby enhancing the image clarity and detail representation. The research adopts a joint learning strategy for the primary and residual dictionaries, enabling the algorithm to progressively and precisely restore high-frequency details in the image, leading to more explicit images after super-resolution processing. Subsequently, the super-resolution operator is compressed using entropy encoding technology, achieving lossless compression for the entire dataset. This improves the visual quality of the images and effectively reduces the storage space required for the dataset. Experimental results demonstrate that the method proposed in this paper achieves a higher compression ratio than traditional lossless compression techniques, with Bps indicators reaching 5.12 and 4.24 and compression times of 0.041 and 0.035 seconds, respectively, while maintaining image quality without degradation. This means that images compressed using this algorithm and uploaded to the cloud can retain their original details and clarity after decompression, providing high-quality image data for subsequent processing. Overall, this achievement offers a new and effective image-processing tool for monitoring milling tool wear in unmanned factories. Leveraging efficient lossless compression technology enables rapid upload of on-site images and cloud processing, reducing computational requirements at the edge. This is of great significance for improving production efficiency and ensuring manufacturing quality.