基于PCNN的面粉麸星检测方法
文献类型：期刊论文
| 作者 | 陈天飞 ; 吴翔; 刘楠嶓; 李秀娟
|
| 刊名 | 中国粮油学报
 |
| 出版日期 | 2015 |
| 卷号 | 30期号:12页码:136-139 |
| 关键词 | 视觉检测 面粉麸星 灰度熵变换 PCNN |
| ISSN号 | 1003-0174 |
| 其他题名 | Bran Specks Detection Approach Based on PCNN |
| 产权排序 | 1 |
| 中文摘要 | 面粉加工过程中麸星数目的多少直接影响着面粉的品质等级,为此,本研究提出了一种基于脉冲耦合神经网络(PCNN)的图像处理方法实现对面粉中微小麸星的视觉检测。首先,该方法对采集的面粉图像进行局部灰度熵变换并通过比例映射生成熵值图像,从而完成了原始面粉图像的图像增强。然后,在图像增强的基础上,利用PCNN对熵值图像进行迭代处理,并通过最小交叉熵确定最优迭代次数,完成最终的麸星目标分割。最后试验验证了该方法的有效性,对比结果表明该方法的检测灵敏度提高近2倍,且算法运行时间为5.189 3s,具有较高的执行效率。 |
| 英文摘要 | Due to the fact that the number of bran specks has directly influence on the flour quality level，so a novel detection approach based on pulse coupled neural network ( PCNN) was proposed so as to achieve vision detection for tiny bran specks in flour．Firstly，the local grey scale entropy transformation was carried on for the original captured image，and the entropy image was formed through proportional mapping，so the bran specks in flour could be enhanced in the entropy image．Secondly，on the basis of image enhancement，the PCNN was applied on the entropy image，and final segmentation results can be obtained after iterative processing，while the optimal number of iterations was determined by the minimum cross entropy．Finally，the actual experimental results demonstrated that the proposed method was effective，and its detection sensitivity was improved nearly twice．In addition，the method runs for 5． 189 3s，and has better execution efficiency． |
| 收录类别 | EI ; CSCD |
| 语种 | 中文 |
| CSCD记录号 | CSCD:5606378 |
| 源URL | [http://ir.sia.cn/handle/173321/17625]  |
| 专题 | 沈阳自动化研究所_装备制造技术研究室 |
| 推荐引用方式 GB/T 7714 | 陈天飞,吴翔,刘楠嶓,等. 基于PCNN的面粉麸星检测方法[J]. 中国粮油学报,2015,30(12):136-139. |
| APA | 陈天飞,吴翔,刘楠嶓,&李秀娟.(2015).基于PCNN的面粉麸星检测方法.中国粮油学报,30(12),136-139. |
| MLA | 陈天飞,et al."基于PCNN的面粉麸星检测方法".中国粮油学报 30.12(2015):136-139. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。