[1]王万亮,江高飞,严江伟,等.基于卷积评价及对抗网络的花粉、孢子图像增广算法[J].江苏农业学报,2021,(05):1190-1198.[doi:doi:10.3969/j.issn.1000-4440.2021.05.014]
　WANG Wan-liang,JIANG Gao-fei,YAN Jiang-wei,et al.Augmented algorithm for pollen and spore images based on convolution evaluation and pix2pix network[J].,2021,(05):1190-1198.[doi:doi:10.3969/j.issn.1000-4440.2021.05.014]
点击复制

基于卷积评价及对抗网络的花粉、孢子图像增广算法()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2021年05期
页码:
1190-1198
栏目:
耕作栽培·资源环境
出版日期:
2021-10-30

文章信息/Info

Title:
Augmented algorithm for pollen and spore images based on convolution evaluation and pix2pix network
作者:
王万亮1; 江高飞2; 严江伟3; 薛卫1
(1.南京农业大学人工智能学院，江苏南京210095;2.南京农业大学资源与环境学院，江苏南京210095;3.山西医科大学法医学院，山西太原030001)
Author(s):
WANG Wan-liang1; JIANG Gao-fei2; YAN Jiang-wei3; XUE Wei1
(1.School of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210095, China;2.College of Resources and Environmental Sciences, Nanjing Agricultural University, Nanjing 210095, China;3.School of Forensic Medicine, Shanxi Medical University, Taiyuan 030001, China)
关键词:
花粉检测; 图像增广; 图像分割; pix2pix网络
Keywords:
pollen detection; image augmentation; image segmentation; pix2pix network
分类号:
S126
DOI:
doi:10.3969/j.issn.1000-4440.2021.05.014
文献标志码:
A
摘要:
针对花粉、孢子图像特征复杂，样本稀缺及种类繁多制约图像检测识别效果的问题，建立基于自适应阈值分割的pix2pix图像增广模型。首先基于卷积评价改进自适应阈值分割算法，择优选取语义分割图像；其次构建pix2pix图像增广模型，将语义分割图像和原始图像建立标签映射用于模型训练，根据语义分割图像生成仿真图像，扩充样本数据集。结果表明，以149种花粉、孢子图像为样本，通过图像增广模型生成的花粉、孢子图像整体相似度达到85.40%；图像增广前Faster RCNN、YOLOv3检测模型的检测精准率分别为86.18%、85.64%，使用增广后的样本训练模型，检测精准率分别达到92.16%、90.57%，提升5.98个百分点和4.93个百分点。
Abstract:
Aiming at the problems such as complex image features of pollens and spores, scarcity of image samples, restricted detection and recognition effects of various images of pollens and spores, a pix2pix image augmentation model based on adaptive threshold segmentation was built. Firstly, the adaptive threshold segmentation algorithm was improved based on convolution evaluation to select the optimal semantic segmentation images of pollens and spores. Secondly, the pix2pix image augmentation model was constructed, the semantic segmentation images and the original images were used to establish label mapping for model training, and emulational pollen and spore images were generated based on semantic segmentation images to extend sample dataset. The results showed that, the overall similarity of 149 pollen and spore images generated by the image augmentation model reached 85.40%. Before image enlargement, the detection accuracies of Faster RCNN and YOLOv3 detection models were 86.18% and 85.64%, respectively. After using the enlarged sample training model, the accuracies reached 92.16% and 90.57%, which increased by 5.98 and 4.93 percentages.

参考文献/References:

[1]吴瑞民，李朝献，李妤蓉. 5种快速方法检测真菌孢子的比较[J]. 国际检验医学杂志,2015,36(9):1257-1258.
[2]刘惠，冀荣华，祁力钧，等. 基于PCA和BP神经网络的万寿菊黑斑病孢子识别[J].中国农业大学学报,2015,20(6):263-268.
[3]谢永华，赵贤国，王琢. 基于突变点几何约束直方图的花粉图像识别[J]. 计算机仿真, 2018,35(12):411-416.
[4]GEETHA N K, BRIDJESH P. Overview of machine learning and its adaptability in mechanical engineering[J]. Materials Today: Proceedings,2020,9:611-617.
[5]RONG C. A summary of research on deep learning in time series learning methods[J]. International Core Journal of Engineering, 2020, 6(9):24-30.
[6]LECROSNIER L, KHEMMAR R, RAGOT N, et al. Deep learning-based object detection, localisation and tracking for smart wheelchair healthcare mobility[J]. International Journal of Environmental Research and Public Health,2020, 18(1):91-99.
[7]YANG J, LI S B, WANG Z, et al. Using deep learning to detect defects in manufacturing: A comprehensive survey and current challenges[J]. Materials (Basel, Switzerland),2020, 13(24):55-57.
[8]MING-FONG T, HUNG-JU T. Enhancing the identification accuracy of deep learning object detection using natural language processing[J]. The Journal of Supercomputing,2021,20(3):525-527.
[9]ITO S, ANDO K, KOBAYASHI K, et al. Automated detection of spinal schwannomas utilizing deep learning based on object detection from magnetic resonance imaging[J]. Spine, 2021, 46(2):195-200.
[10]ZENG Y N, DUAN Q W, CHEN X R, et al. UAVData: A dataset for unmanned aerial vehicle detection[J]. Soft Computing,2021, 25:5385-5393.
[11]HU J Y, RICHARD S C J, ZHANG J S. Saliency-based YOLO for single target detection[J]. Knowledge and Information Systems,2021,63 :717-732.
[12]KARIM S, ZHANG Y, YIN S L, et al. Auxiliary bounding box regression for object detection in optical remote sensing imagery[J]. Sensing and Imaging,2021, 22(1):305-315.
[13]SONG Z N, SUI H G, LI H. A hierarchical object detection method in large-scale optical remote sensing satellite imagery using saliency detection and CNN[J]. International Journal of Remote Sensing, 2021, 42(8):2827-2847.
[14]ZHOU S R, JIA Q. Enhanced SSD with interactive multi-scale attention features for object detection[J]. Multimedia Tools and Applications,2021,80(11):539-556.
[15]LI H, DONG Y, XU L W, et al. Object detection method based on global feature augmentation and adaptive regression in IoT[J]. Neural Computing and Applications,2021, 33(4):119-131.
[16]OMARA I, HAGAG A, MA G Z, et al. A novel approach for ear recognition: learning Mahalanobis distance features from deep CNNs[J]. Machine Vision and Applications,2021, 38(20):1155-1160.
[17]GAO J F, QUMMAR S, ZHANG J M, et al. Ensemble framework of deep CNNs for diabetic retinopathy detection[J]. Computational Intelligence and Neuroscience, 2020,8(6):698-709.
[18]RAMON G C,CARLOS J G O, ANTONIO G M, et al. Precise pollen grain detection in bright field microscopy using deep learning techniques [J].Sensors, 2019, 19(3): 583-595.
[19]VOACTOR S, JOSE A L, AZNARTE I D. Improving classification of pollen grain images of the POLEN23E dataset through three different applications of deep learning convolutional neural networks[J]. PLoS One,2018,13(9): 107-120.
[20]潘浩晨，舒军武. 第四纪主要孢粉数据库评述[J]. 微体古生物学报,2020,37(2):188-196.
[21]PANG Y T, ZHOU X Y, HE W, et al. Uniform design-based gaussian process regression for data-driven rapid fragility assessment of bridges[J]. Journal of Structural Engineering,2021, 147(4):387-399.
[22]WANG J H, GUO Y, WANG Z H, et al. Advancing graph convolution network with revised laplacian matrix[J]. Chinese Journal of Electronics,2020, 29(6):1134-1140.
[23]ZOU L, SONG L T, WEISE T, et al. A survey on regional level set image segmentation models based on the energy functional similarity measure[J]. Neurocomputing, 2020,45(2):606-622.
[24]HUANG C Y, LI X R, WEN Y L. An OTSU image segmentation based on fruitfly optimization algorithm[J]. Alexandria Engineering Journal,2021, 60(1):183-188.
[25]JIANG Y K, QIAN J H, LU S J, et al. LRVRG: a local region-based variational region growing algorithm for fast mandible segmentation from CBCT images[J]. Oral Radiology,2021, 9(1):201-210.
[26]HUANG C, JIN W, XU Q, et al. Sub-pixel edge detection algorithm based on canny-zernike moment method[J]. Journal of Circuits, Systems and Computers,2020, 29(15):273-285.
[27]WANG T, WANG X F, GEORGE S L, et al. Design and analysis of biomarker-integrated clinical trials with adaptive threshold detection and flexible patient enrichment[J]. Journal of Biopharmaceutical Statistics, 2020,30(6) :1060-1076.
[28]WEN P, CARLOS T V, MICHAEL J P. Stochastic pix2pix: A new machine learning method for geophysical and well conditioning of rule-based channel reservoir models[J]. Natural Resources Research, 2020, 30:1319-1345.
[29]张偌雅,李珍珍.数字图像质量评价综述[J].现代计算机,2017(29):78-81.
[30]MA S L, HUANG Y, CHE X J, et al. Faster RCNN-based detection of cervical spinal cord injury and disc degeneration[J]. Journal of Applied Clinical Medical Physics,2020, 21(9):235-243.
[31]AHMED I, JEON G, CHEHRI A, et al. Adapting Gaussian YOLOv3 with transfer learning for overhead view human detection in smart cities and societies[J]. Sustainable Cities and Society, 2021, 10(70):290-298.
[32]HE F, CHUN R K M, QIU Z C, et al. Choroid segmentation of retinal OCT images based on CNN classifier and fitter[J]. Computational and Mathematical Methods in Medicine,2021,33(5):108-119.
[33]ZHANG Z P, HE J, ZHANG Z W, et al. Emotion recognition algorithm based on panorama-plane mapping dataset and VGG16 in prison monitoring system[J]. Journal of Physics: Conference Series,2020, 16(27):120-130.

备注/Memo

备注/Memo:
收稿日期：2021-02-08基金项目：国家自然科学基金重点研发计划项目(2017YFD0800204)作者简介：王万亮(1989-)，男，江苏盐城人，硕士研究生，研究方向为人工智能、图像处理、大数据。(E-mail)superherowang@126.com通讯作者：薛卫， (E-mail)xwsky@njau.edu.cn
更新日期/Last Update: 2021-11-09