Team:UESTC-China/Model
Residual ink counting statistical method
When we were doing our enzymes deinking efficiency validation experiments, we found it is not easy to counting how much ink is residual in the paper that we treated with enzymes. The key problem hindering the counting is that the whole character can't
always completely deinked. This made us didn't measure how much ink was removed. In order to solve this problem, we create an ink counting statistic method based on shrinking characters to dots. It helps
us define how much the residual ink in the paper is simply and directly.
Computer Vision to Complete Ink Recognition
It's easy to tell how many words and pictures there are on a piece of paper. But it's hard to tell how much ink is on the surface of a piece of paper. If we use words as a yardstick, the amount of ink on the picture is hard to evaluate. If we use pictures
as a standard, words are hard to measure. So we need to design a uniform criteria that apply to ink on any form of paper. Then we built “Computer Vision to Complete Ink Recognition” model.
Protein Structure Prediction and Molecular Docking
In the experiment, we genetically modified several of the enzymes we used, adding the Linker and Dockerin domains. But after adding the new part, will the activity of the original enzyme be affected? We need to prove that there is no problem in theoretical dimension with our genetic modification. Second, do we design scaffoldin with steric hindrance after attaching multiple enzymes? This is also a problem that needs to be solved. The key to solve these two problems is
to construct the spatial structure of protein, so that we can intuitively and efficiently demonstrate the conclusion.
Location Decision-making Model of Deinkers
Trash cans, shared bikes, and bus stops are everywhere in the city. But their locations were not chosen at random. The selection of the location of an item depends on the attributes of the item and the attributes of the deployment area. Bike-sharing,
for example, is more common at subway entrances and densely populated areas than in inaccessible mountains. Therefore, it is indispensable to consider the distribution of our deinkers. We need a variety of
data to enrich the properties of the deinker and the region where it will be deployed.