Search

The task of generating sequence images from paragraphs by generating confrontation networks can already generate higher quality images.However,when the input text involves multiple objects and relationships,the context information of the text sequence is difficult to extract,the object layout of the generated image is prone to confusion,and the generated object details are insufficient.To solve this problem,this paper proposes a method of generating sequence images based on scene graphs based on StoryGAN.First,the paragraph is converted into multiple scene graphs through graph convolution,each scene graph contains the object and relationship information of the corresponding text.Then,the bounding box and segmentation mask of the object are predicted to calculate the scene layout.Finally,according to the scene layout and the context information,a sequence of images more in line with the object and its relationship is generated.Tests on CLEVR-SV and CoDraw-SV data sets show that the me-thod in this paper can generate 64×64-pixel sequence images containing multiple objects and their relationships.Experimental results show that on the CLEVR-SV data set,the SSIM and FID of this method are improved by 1.34% and 9.49% respectively than StoryGAN.On the CoDraw-SV data set,the ACC of this method is 7.40% higher than that of StoryGAN.The proposed method improves the rationality of the layout of the generated scene,not only can generate an image sequence containing multiple objects and relationships,but also the generated image has higher quality and clearer details.

This study identified the generation and emission nodes of volatile organic compounds (VOCs) in the metal packaging industry, analyzed the VOCs concentration and species from different production processes, and accounted for secondary pollution through the maximum incremental reactivity method and modified fractional aerosol coefficient method. The results indicated that the main VOCs species were benzenes, alcohols, ketones, and esters, and the benzenes and alcohols contributed more in different types of processes and emission nodes, whereas the ketones and esters contributed less. The maximum concentration was 269.08mg·m -3 ( n -butanol). Strong correlation was found between the concentrations of the production line and their corresponding exhaust, but the VOC species were not totally identical. Furthermore, the potential formations of ozone and secondary organic aerosols were (3.09±0.94) g·g -1 and (2.58±1.99) g·g -1 , respectively, expressed by O 3 /VOCs and SOA/VOCs, and the benzenes and internal coating drying being the major precursors and emission node.

Tibetan red deer (Cervus wallichii) is an endemic species to China, which was once considered extinct in the wild. As there are several other wild ungulates and domestic animals with similar feeding habits within its habitat range, it's thus essential to study interspecific competition and co-existence between Tibetan red deer and other cohabiting ungulates in the highly unique environment of Qinghai-Tibet Plateau. Using microscopic analysis on fresh fecal samples collected in Sangri Tibetan Red Deer Nature Reserve from August to September in 2013 and 2014, the trophic niche width and overlap index were calculated on the basis of diet composition of C. wallichii, Cervus albirostris, Procapra picticaudata, Bos mutus and Capra hircas in green grass period. We analyzed and compared the overlap and differentiation of feeding habits between Tibetan red deer and other wild ungulates and domestic animals. The results showed that C. wallichii fed on similar edible plants with other species, but differed in proportion of different dietary components, with the main edible plants of C. wallichii being mostly the secondary edible plants to other species. Leontopodium pusillum was the common main edible plant for C. wallichii (percentage in animal recipes was 11.2%) and B. mutus (10.2%), Salix xizangensis was the common main edible plant of C. wallichii (9.6%) and C. albirostris (11.4%). At plant family level, Leguminosae was the common main edible plant family for C. wallichii (21.4%) and P. picticaudata (42.5%). Cyperaceae was the common main edible plant family for C. albirostris (49.2%), B. mutus (33.4%) and C. hircas (50.3%). Compositae was main edible plant family for C. wallichii (29.6%), as well as the secondary edible plant family for C. albirostris (7.6%), P. picticaudata (11.6%), B. mutus (17.3%) and C. hircas (14.1%). As the secondary edible plant family for C. wallichii (7.1%), Gramineae took up a lower proportion than that of the other ungulates (C. albirostris (13.6%), P. picticaudata (12.3%), B. mutus (11.5%) and C. hircas (16.0%)). Food overlap indices between C. wallichii and the other ungulates were all higher than 0.5, and the highest with B. mutus (0.65). The food diversity index (1.32), evenness index (0.37) and niche width index (15.79) of C. wallichii were all at high values. Compared with the results from 2007 to 2008, dietary composition of Tibetan red deer changed greatly as the proportion of Leguminosae increased while that of Cyperaceae decreased, resulting in improvement of food quality. In addition, there was greater competition of food resources between C. wallichii and domestic animals, which would further affect the distribution range and living space of C. wallichii.