Image aesthetic quality evaluation using convolution neural network embedded learning

A way of embedded learning convolution neural network(ELCNN) based on the image content is proposed to evaluate the image aesthetic quality in this paper. Our approach can not only solve the problem of small-scale data but also score the image aesthetic quality. First, we chose Alexnet and VGG_S to compare for confirming which is more suitable for this image aesthetic quality evaluation task. Second, to further boost the image aesthetic quality classification performance, we employ the image content to train aesthetic quality classification models. But the training samples become smaller and only using once fine-tuning cannot make full use of the small-scale data set. Third, to solve the problem in second step, a way of using twice fine-tuning continually based on the aesthetic quality label and content label respective is proposed, the classification probability of the trained CNN models is used to evaluate the image aesthetic quality. The experiments are carried on the small-scale data set of Photo Quality. The experiment results show that the classification accuracy rates of our approach are higher than the existing image aesthetic quality evaluation approaches.

Exploring expressive and functional capacities of knitted textiles exposed to wind influence

This study explores the design possibilities with knitted architectural textiles subjected to wind.The purpose is to investigate how such textiles could be applied to alter the usual static expression of exterior architectural and urban elements,such as facades and windbreaks.The design investigations were made on a manual knitting machine and on a CNC(computer numerically controlled)flat knitting machine.Four knitting techniques-tuck stitch,hanging stitches,false lace,and drop stitch-were explored based on their ability to create a three-dimensional effect on the surface level as well as on an architectural scale.Physical textile samples produced using those four techniques were subjected to controlled action of airflow.Digital experiments were also conducted,to probe the possibilities of digitally simulating textile behaviours in wind.The results indicate that especially the drop stitch technique exhibits interesting potentials.The variations in the drop stitch pattern generate both an aesthetic effect of volumetric expression of the textile architectural surface and seem beneficial in terms of wind speed reduction.Thus,these types of knitted textiles could be applied to design architecture that are efficient in terms of improving the aesthetic user experience and comfort in windy urban areas.