油指纹快速识别关键技术研究

发布时间：2018-07-31 20:34

【摘要】：随着我国海洋经济的迅速发展,可持续开发利用海洋资源、切实保护海洋生态环境,防御和减轻海洋灾害,对于提高海洋资源后续保障能力、促进海洋经济健康稳步发展具有重要的战略意义。经济建设的快速发展需求推动着近海石油工业迅速崛起和航运事业的蓬勃发展,我国的原油对外依存度持续攀升,2009年高达51.29%,超过了50%的红线,一举成为世界第二大原油进口国。日益增长的能源需求加大了船舶和油气田发生溢油污染的风险。据不完全统计,从1973年到2006年,我国沿海区域累计发生船舶溢油事件2635起,溢油总量高达37万吨。海洋溢油事件多是突发性的,多发生在恶劣的海洋环境下。虽然多为突发性的,但是一旦发生,危害性却极大。对海洋生态环境造成极大的破坏,给受灾沿岸的经济发展、人类健康、公共安全带来难以估量的危害。鉴于此,本文以油指纹自动识别系统为载体,研究油指纹的不变特征提取算法及自适应多尺度快速识别方法,理论联系实践,优化识别算法,完善油指纹计算机自动识别系统,提高系统的识别精度和鲁棒性,快速准确地确定溢油品种及其来源,为维护国家海洋权益和降低海洋溢油污染提供科技支撑。
[Abstract]:With the rapid development of marine economy in China, the sustainable development and utilization of marine resources, the effective protection of the marine ecological environment, the prevention and mitigation of marine disasters, the enhancement of the ability of marine resources follow-up support, It is of great strategic significance to promote the healthy and steady development of marine economy. The rapid development demand of economic construction has promoted the rapid rise of offshore oil industry and the vigorous development of shipping industry. Our country's crude oil dependence on foreign trade continues to climb, reaching 51.29% in 2009, exceeding 50% of the red line, making China the second largest crude oil importing country in the world. Increasing energy demand increases the risk of oil spills from ships and oil and gas fields. According to incomplete statistics, from 1973 to 2006, there were 2635 incidents of ship oil spill in coastal areas of China, the total amount of oil spill was as high as 370000 tons. Marine oil spills are often sudden and often occur in harsh marine environments. Although most are sudden, but once they occur, the harm is great. It causes great damage to the marine ecological environment, and brings incalculable harm to the economic development, human health and public safety along the affected coast. In view of this, this paper takes the oil fingerprint automatic identification system as the carrier, studies the oil fingerprint invariant feature extraction algorithm and the adaptive multi-scale fast identification method, connects the theory with the practice, optimizes the identification algorithm, consummates the oil fingerprint computer automatic identification system. The identification accuracy and robustness of the system are improved, and the oil spill species and their sources are determined quickly and accurately, which provides scientific and technological support for safeguarding the national marine rights and interests and reducing marine oil spill pollution.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：X55;U698.7

【相似文献】