论文部分内容阅读
在化学结构图中,拐点信息异常重要,如果能对化学结构的拐点进行有效的判别,就可使化学键信息提取的有效性得到很大改善。本研究通过对化学结构的图形分析及特征总结,权衡了Hough变换等提取方法的优势和不足,发现了结构图边缘点偏移值的变化具有很强的规律性,其对图形拐点的提取以及化学键键型的判断有着重要作用。基于以上分析,提出了利用化学结构图偏移值变化的规律性进行化学键信息提取的方法。为了论证算法的有效性,实验时选取了100幅化学结构图,对100幅图形的化学键信息进行了提取并进行了统计分析,得出化学键信息的正确提取率为73.83%,从而通过实验数据证实了算法的可行性。
In the chemical structure diagram, inflection point information is very important. If the inflection point of chemical structure can be discriminated effectively, the effectiveness of chemical bond information extraction can be greatly improved. In this study, by analyzing the chemical structure and summarizing the features, the advantages and disadvantages of the Hough transform and other extraction methods are weighed out. It is found that the variation of the edge point offset value of the structure graph has a strong regularity. The judgment of chemical bond type plays an important role. Based on the above analysis, a method of chemical bond information extraction using the regularity of the change of chemical structure map offset value is proposed. In order to demonstrate the effectiveness of the algorithm, 100 chemical structure maps were selected during the experiment. The chemical bond information of 100 patterns was extracted and statistically analyzed. The correct extraction rate of chemical bond information was 73.83%, which was confirmed by the experimental data The feasibility of the algorithm.