Search

Bit-depth enhancement (BDE) is a challenging task due to stubborn false contour artifacts and disappeared detailed information. Given the mixture of structural distortions and real edges in low bit-depth (LBD) images, both large and small receptive fields (RFs) are critical for BDE tasks. However, even powerful state-of-the-art CNN-based methods can hardly capture sufficient LBD features under multiple RFs. This paper proposes a residual-guided multiscale fusion network (RMFNet) to explore multiscale features in a residual manner. We find that the shuffling operation provides desired multiscale inputs for effectively distinguishing false contours from real edges without any loss of information. Therefore, we shuffle LBD images to multiple scales and then fully extract residual features under different RFs with corresponding subnets. To facilitate interscale guidance from the global context to the local context, we progressively transfer the encoded residual features between adjacent subnets from top to bottom. We further propose a dual-branch depthwise group fusion (DDGF) module to fully capture inter- and inner correlations of multiscale features with fewer parameters. Finally, extensive experiments show that our algorithm achieves excellent performance improvement both quantitatively and qualitatively, verifying its effectiveness. [ABSTRACT FROM AUTHOR]

Our research objective is to lower intralogistics costs by minimizing the number of shuffling operations in a steel plant company commercial warehouse. The process of dispatching products consists of retrieving set of steel bar (SSB) from a floor stored stack or a special stacking frame by an overhead crane. To retrieve a targeted merchandise all SSB above targeted must be reshuffled. Proper assignment of storage locations is a key logistics problem for efficient order picking. We are comparing two heuristics, that do not require information of dispatching sequence of any stored products. We simulated the problem at hand with both methods. Our objective is to count the number of reshuffles using each heuristic on randomly generated examples and decide which is better in the long run. Our problem has similarities with storage assignment of steel plates or steel coils for minimization of reshuffling operations. The problem is also comparable to storage assignment of containers in a container yard. In our case we are dealing with a special stacking configuration of products, that demands different approach. We want to demonstrate which heuristic should be used in companies that lack necessary storage information infrastructure.