[1]温志强,冯向萍,徐 静.基于 Mask R-CNN 的马匹面部别征识别及分割方法[J].计算机技术与发展,2021,31(06):209-215.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 037]
　WEN Zhi-qiang,FENG Xiang-ping,XU Jing.Recognition and Segmentation of Horse Facial FeaturesBased on Mask R-CNN[J].,2021,31(06):209-215.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 037]
点击复制

基于 Mask R-CNN 的马匹面部别征识别及分割方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年06期
页码:
209-215
栏目:
应用前沿与综合
出版日期:
2021-06-10

文章信息/Info

Title:
Recognition and Segmentation of Horse Facial FeaturesBased on Mask R-CNN
文章编号:
1673-629X(2021)06-0209-05
作者:
温志强; 冯向萍; 徐 静
新疆农业大学 计算机与信息工程学院,新疆 乌鲁木齐 830052
Author(s):
WEN Zhi-qiang; FENG Xiang-ping; XU Jing
School of Computer and Information Engineering,Xinjiang Agricultural University,Urumqi 830052,China
关键词:
马匹面部特征; Mask R-CNN; RoI; RolAlign; 特征金字塔; 全卷积网络
Keywords:
facial features of horses; Mask R-CNN; RoI; RolAlign; FPN; FCN
分类号:
TP183;TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 06. 037
摘要:
在马匹鉴定及马匹护照生成的过程中,需要对马匹面部的特征进行文字描述和描图标记。 存在工作量大、效率低、准确性较低、人工成本较高等问题。 为解决这些问题,提出了一种基于 Mask R-CNN 框架对马匹面部及马匹面部特征识别并进行实例分割的方法。 首先通过网络爬虫和马场实地拍摄获取马脸正视图像,利用 Labelme 制作数据集标签,通过ResNet-50-FPN 的主干网络从特征金字塔的不同级别提取 RoI 特征,其中的 RolAlign 算法,使用双线性内插的方法,提高了 RoI 定位精度,最后通过 RoI 单独应用的边界框识别和掩码预测,生成相应的马匹面部及马匹面部特征掩码,实现图像中马匹面部及马匹面部特征与背景的分割。 此外,构建了一个具有分割标注信息的马脸数据集,用于训练相应模型。 实验结果表明,该方法具有较好的马脸检测效果,并能在准确检测的同时实现像素级的马匹面部及马匹面部特征信息分割,可为马匹鉴定和护照生成提供一定支持。
Abstract:
In the process of horse identification and horse passport generation,it is necessary to describe and mark the characteristics of the horse’s face. There are problems such as large workload,low efficiency,low accuracy? and high labor cost. Therefore,we propose a method based on Mask R - CNN framework for horse face recognition and horse facial feature recognition and instance segmentation.Firstly,the face image of the horse is obtained through web crawlers and field shooting,data set labels are made by Label me,and RoIfeatures are extracted from different levels of the feature pyramid through the backbone network of ResNet-50-FPN. Among them,the RollAlign algorithm uses bilinear internal interpolation method to improve the accuracy of RoI positioning. Finally,the corresponding mask of horse face and horse facial features is generated through the boundary box recognition and mask prediction separately applied by RoI to realize the segmentation of horse face and horse facial features from the background in the image. In addition,a horse face dataset with segmented labeling information is constructed for training the corresponding model. The experiment shows that the proposed method has better horse face detection effect and can achieve pixel - level horse face and horse facial feature information segmentation while accurately detecting,which can provide support for horse identification and passport generation.

相似文献/References:

[1]姜世浩,齐苏敏,王来花,等.基于 Mask R-CNN 和多特征融合的实例分割[J].计算机技术与发展,2020,30(09):65.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 012]
　JIANG Shi-hao,QI Su-min,WANG Lai-hua,et al.Instance Segmentation Modal Based on Mask R-CNN and Multi-feature Fusion[J].,2020,30(06):65.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 012]
[2]苏 萍,朱晓辉.基于单目视觉的水面目标识别与测距方法研究[J].计算机技术与发展,2021,31(02):80.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 015]
　SU Ping,ZHU Xiao-hui.Research on Water Surface Target Recognition and Ranging Based on Monocular Vision[J].,2021,31(06):80.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 015]
[3]胡 强,张太红,赵昀杰,等.基于多模型联合的身份证人脸验证应用研究[J].计算机技术与发展,2021,31(08):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 036]
　HU Qiang,ZHANG Tai-hong,ZHAO Yun-jie,et al.Research on Application of ID Face Authentication Based onMulti-model Union[J].,2021,31(06):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 036]
[4]胡一凡,肖满生,范明凯,等.基于 Mask R-CNN 的胃肠息肉图像增强检测[J].计算机技术与发展,2023,33(03):173.[doi:10. 3969 / j. issn. 1673-629X. 2023. 03. 026]
　HU Yi-fan,XIAO Man-sheng,FAN Ming-kai,et al.Mask R-CNN Based Image Enhancement Detection of Gastrointestinal Polyps[J].,2023,33(06):173.[doi:10. 3969 / j. issn. 1673-629X. 2023. 03. 026]

更新日期/Last Update: 2021-06-10