Abstract: An image classification method based on reliable weighted optimal transport (RWOT) includes: preprocessing data in a source domain, so that a deep neural network fits a sample image in the source domain to obtain a sample label; performing image labeling to add a pseudo label to a data sample in a target domain; performing node pairing to pair associated images in the source domain and the target domain; and performing automatic analysis by using a feature extractor and an adaptive discriminator, to perform image classification. The present disclosure proposes a subspace reliability method for dynamically measuring a difference between the source domain and the target domain based on spatial prototypical information and an intra-domain structure. This method can be used as a preprocessing step of an existing domain adaptation technology, and greatly improves efficiency.

Abstract: This invention relates to a system for recuperation and utilization of exhaust heat from a reformer, in which a plurality of arc passageways are formed respectively on the opposite outside walls of the reformer body with a plurality of arc-shaped steel sheet or zinc-plated steel sheet, said arch passageways together with the bottom chamber under the reformer body serve as preheating duct, and at least one air inlet above each passageway is disposed above the top of the reformer. Under the effect of a blower, the air above the reformer top passes through the said preheating duct, and absorbs the heat dispersed from the reformer body. The preheated air makes indirect heat exchange with the flue gas through a heat pipe excharger. The temperature of the air after heat exchange rises from the environmental to about 220.degree. C. and the air is then introduced to the reformer and auxiliary boiler for combustion. The temperature of the flue gas drops from about 300.degree. C. to about 160.degree. C.