| 许枫,张乔,张纯,苏瑞文.Walsh变换对鱼类特征识别的研究*[J].,2015,34(5):465-472 |
| Walsh变换对鱼类特征识别的研究* |
| Walsh transform for fish identification research |
| 投稿时间：2015-02-06 修订日期：2015-08-26 |
| 中文摘要: |
| 鱼种的快速识别是渔业资源评估乃至海洋生态系统监测重要组成部分。与传统的拖网捕捞等方法相比,声学方法具有快速有效、调查区域广、不损坏生物资源、可持续观察等优点。鱼类的声学识别方法主要是基于鱼类回波信号特征的识别, 鱼体形状及组成结构的复杂多样导致其回波信号非常复杂,因此利用简单的回波包络或能量特征识别鱼类效果往往不能令人满意。本文提出一种基于Walsh 变换的鱼类回波识别方法。试验获取鲫鱼、 嘎鱼、武昌鱼的回波信号,处理过程中分别提取三种鱼类回波包络信号的Walsh谱作为识别特征量,并利用BP神经网络分类器对其进行了分类。结果表明利用回波的Walsh谱可以成功识别不同形状的鱼类,其中对武昌鱼的识别正确率达90%以上。 |
| 英文摘要: |
| The fast and efficient fish identification is the composition of fishery survey and marine ecosystem monitoring. Compared with the trawling way, the acoustics method is efficient, wideness, less invasion and sustainable, which is mostly based on the backscattering of fish. The analysis of fish backscattering is difficult due to the complex structure and variable composition of fish. Consequently, the methods of fish identification are limited. A method of fish identification based on Walsh transform is proposed in this paper. Firstly, an ex situ experiment has been performed with three kinds of fish: Crucian carp (Carassius auratus), Yellow-headed catfish (Pelteobagrus fulvidraco) and Bluntnose black bream (Megalobrama amblycephale). The backscattering signals of these fishes are obtained to verify this method. Then, the Walsh spectrum of backscattering is extracted as the indicator to describe these three kinds of fish species. Finally, three kinds of fish are successfully identified by using a BP neural network. The result shows that it’s possible to identify fish with different shape using Walsh transform. |
| DOI： |
| 中文关键词: Walsh变换,特征提取,鱼类识别,BP神经网络 |
| 英文关键词: Walsh Transform Feature Extraction Fish Identification BP neural network |
| 基金项目:山东省科技发展计划项目 |
|
| 摘要点击次数: 2343 |
| 全文下载次数: 1997 |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
