Machine-learning-based quality-level-estimation system for inspecting steel microstructures.

Quality control of special steel is accomplished through visual inspection of its microstructure based on microscopic images. This study proposes an 'automatic-quality-level-estimation system' based on machine learning. Visual inspection of this type is sensory-based, so training data may include variations in judgments and training errors due to individual differences between inspectors, which makes it easy for a drop in generalization performance to occur due to overfitting. To deal with this issue, we here propose the preprocessing of inspection images and a data augmentation technique. Preprocessing reduces variation in images by extracting features that are highly related to the level of quality from inspection images. Data augmentation, meanwhile, suppresses the problem of overfitting when training with a small number of images by taking into account information on variation in judgment values obtained from on-site experience. While the correct-answer rate for judging the quality level by an inspector was about 90%, the proposed method achieved a correct-answer rate of 92.5%, which indicates that the method shows promise for practical applications.

Glucosylation of acetic acid by sucrose phosphorylase.

Transglucosylation from sucrose to acetic acid by sucrose phosphorylase (EC 2.4.1.7) was studied. 1-O-Acetyl-alpha-D-glucopyranose was isolated as the main product of the enzyme reaction. We also compared the pH-dependence of transglycosylation catalyzed by sucrose phosphorylase toward carboxyl and hydroxyl groups. With hydroquinone as an acceptor molecule, the transfer ratio of glucose residue was higher at neutral pH. This pH-activity profile was similar to that of the phosphorolysis of sucrose by sucrose phosphorylase, but with acetic acid as an acceptor molecule, the transfer ratio of glucose residue was higher at low pH. These findings suggest that the undissociated carboxyl group is essential to the acceptor molecule for the transglycosylation reaction of sucrose phosphorylase. In a sensory test, the sour taste of acetic acid was markedly reduced by glucosylation. The threshold value of the sour taste of acetic acid glucosides was approximately 100 times greater than that of acetic acid.

Novel transglucosylating reaction of sucrose phosphorylase to carboxylic compounds such as benzoic acid.

We examined the synthesis of benzoyl glucoside using the transglucosylation reaction of sucrose phosphorylase. Sucrose phosphorylase from Streptococcus mutans showed marked transglucosylating activity, particularly under acidic conditions. On the other hand, sucrose phosphorylase from Leuconostoc mesenteroides showed very weak transglucosylating activity. Three main products were detected from the reaction mixture using benzoic acid as an acceptor molecule and sucrose as a donor molecule. These compounds were identified as 1-O-benzoyl alpha-D-glucopyranoside, 2-O-benzoyl alpha-D-glucopyranose and 2-O-benzoyl beta-D-glucopyranose on the basis of their isolation and the isolation of their acetylated products and subsequent analysis using 1D- and 2D-NMR analyses. From the results of the time-course analyses of the enzyme reaction and the degradation of 1-O-benzoyl alpha-D-glucopyranoside, 1-O-benzoyl alpha-D-glucopyranoside was considered to be initially produced by the transglucosylation reaction of the enzyme, and 2-O-benzoyl alpha-D-glucopyranose and 2-O-benzoyl beta-D-glucopyranose were produced from 1-O-benzoyl alpha-D-glucopyranoside by intramolecular acyl migration reaction. The acceptor specificity in the glucosylation reaction of S. mutans sucrose phosphorylase was also investigated. This sucrose phosphorylase could transglucosylate toward various carboxylic compounds. Short-chain fatty acids, hydroxy acids and dicarboxylic acids were also glucosylated with this sucrose phosphorylase.