基于计算机视觉的图像多尺度识别方法

doi:10.11772/j.issn.1001-9081.2015.02.0502

计算机应用 ›› 2015, Vol. 35 ›› Issue (2): 502-505.DOI: 10.11772/j.issn.1001-9081.2015.02.0502

基于计算机视觉的图像多尺度识别方法

张玉璞¹, 杨旗¹, 张旗²

1. 沈阳理工大学机械工程学院, 沈阳 110159;
2. 国网辽宁省电力有限公司营口供电公司, 辽宁营口 115000

收稿日期:2014-09-16 修回日期:2014-11-17 出版日期:2015-02-10 发布日期:2015-02-12
通讯作者: 杨旗
作者简介:张玉璞(1975-),男,山东蓬莱人,讲师,硕士,主要研究方向:图像识别、自动控制; 杨旗(1976-),男,辽宁沈阳人,副教授,博士,主要研究方向:步态识别、图像识别; 张旗(1977-),男,辽宁营口人,工程师,硕士,主要研究方向:图像识别、电网控制。
基金资助:
辽宁省科技厅资助项目(20131097);辽宁省教育厅资助项目(L2014082)。

Image multi-scale recognition method based on computer vision

ZHANG Yupu¹, YANG Qi¹, ZHANG Qi²

1. School of Mechanical Engineering, Shenyang Ligong University, Shenyang Liaoning 110159, China;
2. State Grid Yingkou Electric Power Supply Company, Yingkou Liaoning 115000, China

Received:2014-09-16 Revised:2014-11-17 Online:2015-02-10 Published:2015-02-12

摘要/Abstract

摘要：

针对图像识别中图像尺寸比例不一致、旋转角度不相同,以及识别率低、鲁棒性差的问题,提出一种图像的形态学识别算法。首先对图像进行中心化及归一化处理,同时将图形的轮廓图像转换为二值图像;其次采用大小变化的圆形进行图像过滤,获取图形不同尺寸的形态学特征,建立扇形区域特征向量;最后采用多尺度的分析方法进行图像识别以及图像角度分析。在角度无关性、比例无关性、轮廓干扰鲁棒性下与传统方法进行对比实验,实验结果表明该方法有较高的识别率,并可以分析出图像间的角度差值,在图像轮廓有噪声的情况下有较好的鲁棒性,同时大大降低了图像尺寸比例不同、旋转角度不同对图像识别的影响。

关键词: 图像识别, 计算机视觉, 多尺度, 形态学分析, 鲁棒性

Abstract:

Focusing on the issues of size-varying and angle-varying of the images, and low recognition rate and poor robustness in image recognition, a morphological image recognition method was proposed. Firstly, image was centralized and normalized, and the silhouettes of image was converted into binary image. Secondly, varable circles were used to extract morphological features of image, and a fan-shaped area feature vector was established. Finally, multi-scale analysis method was applied to image recognition and image angle analysis. Compared with traditional method in the conditions such as angle independence, proportion independence and profile robustness, the experimental results show that the proposed method has higher recognition rate, and can analyze the angle difference between the images. The method is robust to noise, and can significantly reduce the influence of different image scale and rotation angle on image recognition.

Key words: image recognition, computer vision, multi-scale, morphological analysis, robustness

中图分类号:

TP391.41

张玉璞, 杨旗, 张旗. 基于计算机视觉的图像多尺度识别方法[J]. 计算机应用, 2015, 35(2): 502-505.

ZHANG Yupu, YANG Qi, ZHANG Qi. Image multi-scale recognition method based on computer vision[J]. Journal of Computer Applications, 2015, 35(2): 502-505.

参考文献

[1] CUI J, SONG X, CHEN G, et al. Feature extraction and matching of vein based on geometrical shape and wavelet moment [J]. Journal of Northeastern University: Natural Science Edition, 2009,30(9):1236-1240. (崔建江, 宋星月, 陈国坤,等. 基于几何形状和小波矩的静脉特征提取与匹配[J]. 东北大学学报:自然科学版, 2009,30(9):1236-1240.)
[2] CUI P, PAN Q, LI J, et al. Image recognition using projection-based wavelet moment [J]. Journal of Electronics and Information Technology, 2005,27(11):1774-1777. (崔培玲,潘泉,李军宏,等.一种基于投影的小波矩及其在图像识别中的应用[J].电子与信息学报,2005,27(11):1774-1777.)
[3] LIU Y, YOU Z, CAO L, et al. A multiresolution image recognition method using Fisher transform and its application to car type recognition [J]. Journal of Electronics and Information Technology, 2003,25(12):1603-1611. (刘怡光,游志胜,曹丽萍,等. 基于Fisher变换的多尺度图像识别方法及其车形识别应用[J]. 电子与信息学报,2003,25(12):1603-1611.)
[4] LONG F, DONG H, WANG B, et al. Gabor representation based probabilistic subspace analysis for face recognition [J]. Journal of Electronics and Information Technology, 2007,29(3):626-630. (龙飞,董槐林,王备战,等. 一种基于Gabor描述的概率子空间人脸识别方法[J]. 电子与信息学报,2007,29(3):626-630.)
[5] LIAO L, ZHANG Y, MAYBANKC S J, et al. Image recognition via two-dimensional random projection and nearest constrained subspace [J]. Journal of Visual Communication and Image Representation, 2014, 25(5): 1187-1198
[6] MENG X, WANG Z, WU L. Building global image features for scene recognition [J]. Pattern Recognition, 2012,45(1): 373-380.
[7] LIU Q, LU H, MA S. A survey: subspace analysis for face recognition [J]. Acta Automatica Sinica, 2003,29(6):900-911. (刘青山,卢汉清,马颂德. 综述人脸识别中的子空间方法[J]. 自动化学报,2003,29(6):900-911.)
[8] ZHANG Y. Enhanced statistical analysis of nonlinear processes using KPCA, KICA and SVM [J]. Chemical Engineering Science, 2009, 64(5): 801-811
[9] WU J, WANG J, LIU L. Feature extraction via KPCA for classification of gait patterns [J]. Human Movement Science, 2007, 26(3): 393-411.
[10] ZHAO M, LI S, KWOK J. Text detection in images using sparse representation with discriminative dictionaries [J]. Image and Vision Computing, 2010, 28(12): 1590-1599
[11] NAKHMANI A, TANNENBAUM A. A new distance measure based on generalized image normalized cross-correlation for robust video tracking and image recognition [J]. Pattern Recognition Letters, 2013, 34(3): 315-321.
[12] RAGHAVENDRA R, BUSCH C. Novel image fusion scheme based on dependency measure for robust multispectral palmprint recognition [J]. Pattern Recognition, 2014, 47(6): 2205-2221.
[13] SAVCHENKO A V. Directed enumeration method in image recognition [J]. Pattern Recognition, 2012, 45(8): 2952-2961.
[14] YANG Q, XUE D. Gait recognition based on two-scale dynamic Bayesian network and more information fusion [J]. Journal of Electronics and Information Technology, 2012,34(5):1148-1153. (杨旗, 薛定宇. 基于双尺度动态贝叶斯网络及多信息融合的步态识别[J]. 电子与信息学报,2012,34(5):1148-1153.)
[15] YANG Q, XUE D. Gait recognition based on dynamic & static information fusion and dynamic bayesian network [J].Journal of Image and Graphics, 2012, 17(7):783-790. (杨旗, 薛定宇. 动静态信息融合及动态贝叶斯网络的步态识别[J]. 中国图象图形学报, 2012, 17(7):783-790.)

基于计算机视觉的图像多尺度识别方法

Image multi-scale recognition method based on computer vision

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	高工, 杨红雨, 刘洪. 基于深度学习的三维点云人脸识别[J]. 计算机应用, 2021, 41(9): 2736-2740.
[2]	马佳良, 陈斌, 孙晓飞. 基于改进的Faster R-CNN的通用目标检测框架[J]. 计算机应用, 2021, 41(9): 2712-2719.
[3]	牛康力, 谌雨章, 沈君凤, 曾张帆, 潘永才, 王绎冲. 基于深度学习的双通道夜视图像复原方法[J]. 计算机应用, 2021, 41(6): 1775-1784.
[4]	郭天晓, 胡庆锐, 李建伟, 沈燕飞. 基于人体骨架特征编码的健身动作识别方法[J]. 计算机应用, 2021, 41(5): 1458-1464.
[5]	曹建芳, 田晓东, 贾一鸣, 闫敏敏. 改进DeepLabV3+模型在壁画分割中的应用[J]. 计算机应用, 2021, 41(5): 1471-1476.
[6]	欧莉莉, 邵峰晶, 孙仁诚, 隋毅. 基于半监督方法的脑梗死图像识别[J]. 计算机应用, 2021, 41(4): 1221-1226.
[7]	蒋丽, 黄仕建, 严文娟. 基于低秩行为信息和多尺度卷积神经网络的人体行为识别方法[J]. 计算机应用, 2021, 41(3): 721-726.
[8]	杜培德, 严华. 基于多尺度空间注意力特征融合的人群计数网络[J]. 计算机应用, 2021, 41(2): 537-543.
[9]	裴仪瑶, 郭会明, 张丹普, 陈文博. 基于定位不确定性的鲁棒3D目标检测方法[J]. 计算机应用, 2021, 41(10): 2979-2984.
[10]	韩建栋, 李晓宇. 基于多尺度特征融合的行人重识别方法[J]. 计算机应用, 2021, 41(10): 2991-2996.
[11]	张凯悦, 张鸿. 基于注意力机制网络的航运监控图像识别模型[J]. 计算机应用, 2021, 41(10): 3010-3016.
[12]	肖勇, 郑楷洪, 郑镇境, 钱斌, 李森, 马千里. 基于多尺度跳跃深度长短期记忆网络的短期多变量负荷预测[J]. 计算机应用, 2021, 41(1): 231-236.
[13]	温静, 宋建伟. 基于多级全局信息传递模型的视觉显著性检测[J]. 计算机应用, 2021, 41(1): 208-214.
[14]	高智勇, 黄金镇, 杜程刚. 基于特征金字塔网络的肺结节检测[J]. 计算机应用, 2020, 40(9): 2571-2576.
[15]	张永宏, 刘昊, 田伟, 王剑庚. 基于DeepLab v3的西藏地区降雨云团分割方法[J]. 计算机应用, 2020, 40(9): 2781-2788.