|Table of Contents|

Blind Sidewalk Segmentation Algorithm Based on ImagePreprocessing Classification and Segmentation(PDF)

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

Issue:
2020年01期
Page:
42-48
Research Field:
信息与通信工程
Publishing date:

Info

Title:
Blind Sidewalk Segmentation Algorithm Based on ImagePreprocessing Classification and Segmentation
Author(s):
Liu Ling; Sun Chenchen; Xu Yinlin; Tang Wanchun; Zhao Hua
School of Physics and Technology,Nanjing Normal University,Nanjing 210023,China
Keywords:
preprocessing classification; color histogram; support vector machine(SVM); threshold segmentation
PACS:
TP391.4
DOI:
10.3969/j.issn.1672-1292.2020.01.007
Abstract:
The blind sidewalks object segmentation algorithm with high accurate segmentation rate is an important key to a high performance blind guided system. In this paper,a color histogram support vector machine(SVM)method is proposed to classify blind sidewalks into color blind sidewalks or texture blind sidewalks. For the color blind sidewalks,an improved OTSU segmentation method is proposed for multi-parameter fusion of HSV color space. For the texture blind sidewalks,based on texture enhancement,K-means clustering method is proposed. Because of the effective preprocessing classification method,they can be recognized according to the color or texture features of blind sidewalks. Besides the improved color and texture segmentation algorithm has excellent adaptability for different kinds of blind sidewalks under the different environments,and the average segmentation accuracy of the blind sidewalks can reach more than 90% for the test library pictures.

References:

[1] 耿魁,张春阳,凌海燕. 城市盲道的缺陷及现状分析—以烟台市为例[J]. 科技传播,2014(14):76-77.
[2]葛忠明,张忠海.“健全人的盲道”:关于盲道占用的社会排斥研究[J]. 中国海洋大学学报(社会科学版),2016(1):81-88.
[3]王民,李媛,张立材. 基于纹理特征的盲道区域分割算法[J]. 信息通信,2017(7):23-26.
[4]闫志杰,汪剑鸣,窦汝振,等. 基于颜色聚类和直线检测的自适应盲道区域分割算法[J]. 天津工业大学学报,2010,29(1):80-84.
[5]周毅,赵群飞. 基于颜色信息的盲道区域检测与跟随算法[J]. 微型电脑应用,2010,26(8):47-50.
[6]柯剑光,赵群飞,施鹏飞. 基于图像处理的盲道识别算法[J]. 计算机工程,2009,35(1):189-191,197.
[7]彭玉青,薛杰,郭永芳. 基于颜色纹理信息的盲道识别算法[J]. 计算机应用,2014,34(12):3585-3588,3604.
[8]仲会娟. 基于颜色特征和SVM的自然图像分类标注算法[J]. 绵阳师范学院学报,2018,37(5):12-16.
[9]陈亦欣,叶锋,肖锋,等. 基于HSV空间和形状特征的交通标志检测识别研究[J]. 江汉大学学报(自然科学版),2016,44(2):119-125.
[10]张国宣,孔锐,施泽生,等. 基于核聚类方法的多层次支持向量机分类树[J]. 计算机工程,2005,31(5):172-174.
[11]程石磊,郝佳宁,陈钰. 基于MATLAB带钢表面缺陷检测算法设计[J]. 数字化用户,2018,24(35):233.
[12]程洪,尹辉娟,马丽,等. HSI模型结合OTSU 法的玉米种胚检测算法[J]. 农机化研究,2014(10):183-185,191.
[13]ARISTIDIS L,NIKOS V,JAKOB J V. The global k-means clustering algorithm[J]. Pattern Recognition,2003,36(2):451-461.
[14]TAN X,TRIGGS B. Enhanced local texture feature sets for face recognition under difficult lighting conditions[J]. IEEE Transactions on Image Processing,2010,19(6):1635-1650.
[15]彭志涛,陈风东,唐军,等. 高斯差分滤波多尺度损伤提取方法[J]. 强激光与粒子束,2017,29(9):16-18.
[16]张存. 基于Quick Bird影像的面向对象信息提取方法比较实验研究[D]. 沈阳:东北大学,2009.
[17]何德鲁,李宏坤. 基于改进的灰度共生矩阵参数特征提取的齿轮箱故障诊断[J]. 机械管理开发,2016(9):62-64.
[18]薄华,马缚龙,焦李成. 图像纹理的灰度共生矩阵计算问题的分析[J]. 电子学报,2006,34(1):155-158,134.
[19]郭颖. 基于区域生长的肝脏CT图像肿瘤分割方法研究[D]. 大连:大连海事大学,2017.
[20]秦志光,陈浩,丁熠,等. 基于多模态卷积神经网络的脑血管提取方法研究[J]. 电子科技大学学报,2016,45(4):573-581.
[21]鲁浩达,徐军,刘利卉,等. 基于深度卷积神经网络的肾透明细胞癌细胞核分割[J]. 生物医学工程研究,2017,36(4):340-345.

Memo

Memo:
-
Last Update: 2020-03-15