纺织学报 ›› 2017, Vol. 38 ›› Issue (12): 150-156.doi: 10.13475/20161202907

• 管理与信息化 • 上一篇    下一篇

应用卷积网络及深度学习理论的羊绒与羊毛鉴别

  

  • 收稿日期:2016-12-20 修回日期:2017-08-29 出版日期:2017-12-15 发布日期:2017-12-18

Iedntification of cashmere and wool based on convolutional neuron networks and deep learning theory

  • Received:2016-12-20 Revised:2017-08-29 Online:2017-12-15 Published:2017-12-18

摘要:

为解决羊绒与羊毛纤维图像难以鉴别的问题,提出一种基于卷积网络和深度学习理论的鉴别方法。使用sigmoid分类器将卷积深度网络提取的纤维图像特征进行粗分类,根据验证集合验证结果并记录网络的最优权重。根据整体的分类网络所获取的权值,对每张样本图片使用改进的局部增强整体的网络模型提取局部特征,并对局部特征和整体特征进行融合,根据这些融合特征建立新的分类网络。在此基础上,使用鄂尔多斯标准羊绒与羊毛数据集对网络进行50轮次的迭代训练,得到的最优准确率达92.1%。实验结果表明:采用深度卷积网络表征纤维,并对羊绒羊与毛纤维图像进行分类的方法,能够有效解决羊绒、羊毛等类似纤维鉴别问题;若用于商业检测,还需更多数据集的验证。

关键词: 羊绒, 羊毛, 图像鉴别, 卷积网络

Abstract:

In order to solve the problem of the difficult in identifying of cashmere and wool fiber images, a novel identification method based on convolutional neuron (CNN) and the deep learning theory was proposed. A sigmoid classier was used to carry out coarse classification on fiber image features extracted by the CNN. The reaults were verified according to the validation datasets, and the optimal weights of the network was recorded. Part-level features of each sample image are extracted by an improved part-level augment object-level network based on previously obtained parameters. In addition, the part-level and object-level features were fused, and a new network model was established based on the fused features. On this basis, Ordos’s standard cashmere and wool dataset was used to train the network for 50 times, and the best accuracy is 92.1%. The experiments results demonstrate that the method for classifying cashmere and wool images, based on the CNN features, can be applied to cashmere and wool or discriminate similar fibers identification successfully. However, a large number of sample sets and validations are required for commerical use.

Key words: cashmere, wool, image identification, convolutional neuron networks

[1] 王瑞洁 李龙 秦彩霞 . 采用滑溜牵伸的低比例山羊绒混纺纺纱实践[J]. 纺织学报, 2018, 39(06): 24-28.
[2] 郑君红 李亮 刘让同 张丹. 羊毛角蛋白的制备及其对涤纶织物的整理[J]. 纺织学报, 2018, 39(03): 92-97.
[3] 任燕飞 巩继贤 付冉冉 张健飞 王富邦 陶宇庆. 微生物合成纳米灵菌红素及其对羊毛织物抗菌染色[J]. 纺织学报, 2018, 39(02): 91-96.
[4] 刘建勇 吴胜争 赵笑康. 生物酶协同催化体系及其对羊毛纤维的作用机制[J]. 纺织学报, 2018, 39(01): 71-78.
[5] 薛日杰 王树根 施楣梧. TiO2-SnO2复合溶胶在羊毛阻燃整理中的应用[J]. 纺织学报, 2017, 38(12): 95-100.
[6] 贾丽霞 金崇业 刘瑞 单国华. 硅磷杂化阻燃整理对羊毛结构与热稳定性能的影响[J]. 纺织学报, 2017, 38(12): 101-105.
[7] 陈诚 贾丽霞 张初阳. 毛用防蛀萘醌色素的合成与性能评价[J]. 纺织学报, 2017, 38(10): 70-74.
[8] 朱俊平 路凯 柴新玉 钟跃崎 . 羊绒与羊毛直径的水平集中轴线法测量[J]. 纺织学报, 2017, 38(09): 14-18.
[9] 杨树 李玛莎. 羊毛纤维集合体的分形结构与其保暖性的关系[J]. 纺织学报, 2017, 38(08): 11-15.
[10] 韦玉辉 宁琳 吴雄英 丁雪梅. 家用干衣机滚筒烘干方式对羊毛织物性能的影响[J]. 纺织学报, 2017, 38(07): 69-74.
[11] 刘让同 刘淑萍 李亮 李淑静 田广菓 弋梦梦 朱雪莹. 电子显微镜下羊毛的溶解模式[J]. 纺织学报, 2017, 38(07): 1-5.
[12] 路凯 钟跃崎 朱俊平 柴新玉. 基于视觉词袋模型的羊绒与羊毛快速鉴别方法[J]. 纺织学报, 2017, 38(07): 130-134.
[13] 栾巧丽 邱华 成钢 刘晓燕. 羊毛及其混合纤维非织造材料的吸声性能[J]. 纺织学报, 2017, 38(03): 67-71.
[14] 王宗乾 杨海伟 汤立洋 解项东. 漂白羊毛的防紫外线整理及其耐光稳定性能[J]. 纺织学报, 2017, 38(03): 99-107.
[15] 赵明宇 刘海辉 王学晨 张兴祥. 角蛋白/多壁碳纳米管复合纤维的制备[J]. 纺织学报, 2017, 38(02): 21-25.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!