为了提升多目标自动分割和分类效果，设计了基于模式识别的图像中多目标自动分割和分类方法。利用分水岭算法提取图像中多目标过分割区域；通过模式识别中改进模糊C均值聚类算法，聚类处理提取的多目标过分割区域，得到多目标自动分割结果；在模式识别中的卷积神经网络内，输入自动分割结果，输出多目标自动分类结果。实验证明：该方法可有效获取多目标过分割区域，得到多目标自动分割结果；在不同图像分辨率时，该方法自动分割的划分系数较大、划分熵较小，即自动分割效果较佳；该方法具备多目标自动分类的可行性，且自动分类精度较高。
To improve the effect of automatic segmentation and classification of multiple objects in images based on pattern recognition is studied. Watershed algorithm is used to extract multi-target over-segmented regions in the image. Through the improved fuzzy C-means clustering algorithm in pattern recognition, the extracted multi-target over-segmented regions are clustered, and the multi-target automatic segmentation results are obtained. In the convolutional neural network of pattern recognition, the automatic segmentation results are input and the automatic multi-target classification results are output. Experimental results show that the proposed method can effectively obtain the over-segmented regions of multiple targets and obtain the automatic segmentation results of multiple targets. When the image resolution is different, the partition coefficient is larger and the partition entropy is smaller, that is, the automatic segmentation effect is better. This method has the feasibility of multi-target automatic classification, and the automatic classification accuracy is high.
2022,44(20): 153-156 收稿日期：2022-05-26
DOI：10.3404/j.issn.1672-7649.2022.20.031
分类号：TP751.1
基金项目：黑龙江省哲学社会科学研究规划资助项目(19YYE304)
作者简介：陈峰(1982-)，男，博士，讲师，研究方向为多相流检测、模式识别及计算机视觉
参考文献：
[1] 张文坤, 汪西原, 韩佳雪. 自动确定类别数的RJMCMC+SA图像分割算法研究[J]. 图学学报, 2019, 40(6): 1038–1047
ZHANG Wenkun, WANG Xiyuan, HAN Jiaxue. The research about RJMCMC+SA image segmentation algorithm to automatically determine the number of categories[J]. Journal of Graphics, 2019, 40(6): 1038–1047
[2] 李晖晖, 周康鹏, 韩太初. 基于CReLU和FPN改进的SSD舰船目标检测[J]. 仪器仪表学报, 2020, 41(4): 183–190
LI Huihui, ZHOU Kangpeng, HAN Taichu. Ship object detection based on SSD improved with CReLU and FPN[J]. Chinese Journal of Scientific Instrument, 2020, 41(4): 183–190
[3] 马健, 史文旭, 鲍胜利. 基于特征融合SSD的遥感图像舰船目标检测[J]. 计算机应用, 2019, 39(S2): 253–256
MA Jian, SHI Wenxu, BAO Shengli. Ship target detection in remote sensing images based on feature fusion SSD[J]. Journal of Computer Applications, 2019, 39(S2): 253–256
[4] 丁晓娜, 刘春凤, 刘保相. 基于PCA图像粒化的多粒度图像分类模型研究[J]. 山西大学学报(自然科学版), 2020, 43(4): 706–712
DING Xiaona, LIU Chunfeng, LIU Baoxiang. Research on multi-granularity image classification model based on PCA image Granulation[J]. Journal of Shanxi University (Natural Science Edition), 2020, 43(4): 706–712
[5] 赵凤, 张咪咪, 刘汉强. 区域信息驱动的多目标进化半监督模糊聚类图像分割算法[J]. 电子与信息学报, 2019, 41(5): 1106–1113
ZHAO Feng, ZHANG Mimi, LIU Hanqiang. Multi-objective evolutionary semi-supervised fuzzy clustering image segmentation motivated by region information[J]. Journal of Electronics & Information Technology, 2019, 41(5): 1106–1113
[6] 赵鹏, 汪纯燕, 张思颖, 等. 一种基于融合重构的子空间学习的零样本图像分类方法[J]. 计算机学报, 2021, 44(2): 409–421
ZHAO Peng, WANG Chunyan, ZHANG Siying, et al. A Zero-shot image classification method based on subspace learning with the fusion of reconstruction[J]. Chinese Journal of Computers, 2021, 44(2): 409–421
[7] 付晓, 沈远彤, 李宏伟, 等. 基于半监督编码生成对抗网络的图像分类模型[J]. 自动化学报, 2020, 46(3): 531–539
FU Xiao, SHEN Yuantong, LI Hongwei, et al. A Semi-supervised encoder generative adversarial networks model for image classification[J]. Acta Automatica Sinica, 2020, 46(3): 531–539
[8] 韩哲, 李灯熬, 赵菊敏, 等. 基于改进遗传模糊聚类和水平集的图像分割算法[J]. 计算机工程与设计, 2019, 40(5): 1390–1393+1412
HAN Zhe, LI Deng ao, ZHAO Jumin, et al. Image segmentation algorithm based on improved genetic fuzzy clustering and level set[J]. Computer Engineering and Design, 2019, 40(5): 1390–1393+1412
