Pork Registration Using Skin Image with Deep Neural Network Features

Pork food safety is not optimistic in China. Some pork carrying viruses can damage the liver and kidneys of consumers. At present, the pork traceability system in China mainly relies on the information of the stamp on the pork skin. However, there is no unified stamp standard in China to regulate pork market, and information on the stamp is likely to be destroyed during the circulation process. In this study, deep convolutional neural network (DCNN) was used to extract the features of pork skin, and dynamic inlier selection was used to register pork skin images to achieve accurate traceability of pork. The data set consists of 810 images, which containing images captured from three angles and nine position. The results show that features extracted using DCNN are better than SIFT features, and the average matching rate is 92.59%. Compared with KNN (K-Nearest Neighbor), CPD (Coherent Point Drift), ICP (Iterative Closest Point), our dynamic inlier selection has better registration effect. In our pork skin dataset, the success rate of registration reached 86.67%, which provided a reference for subsequent pork traceability research.

[1]  Francisco Argüello,et al.  GPU Accelerated FFT-Based Registration of Hyperspectral Scenes , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[2]  Max A. Viergever,et al.  A survey of medical image registration , 1998, Medical Image Anal..

[3]  Xiaolei Dong,et al.  PPOIM: Privacy-Preserving Shape Context Based Image Denoising and Matching with Efficient Outsourcing , 2018, ICICS.

[4]  Michael I. Jordan,et al.  Kernel independent component analysis , 2003 .

[5]  David L. Ortega,et al.  Modeling heterogeneity in consumer preferences for select food safety attributes in China , 2011 .

[6]  Behrooz Nasihatkon,et al.  Globally Optimal Rigid Intensity Based Registration: A Fast Fourier Domain Approach , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[8]  Yansheng Li,et al.  Feature guided Gaussian mixture model with semi-supervised EM and local geometric constraint for retinal image registration , 2017, Inf. Sci..

[9]  Xiaotao Huang,et al.  Enhanced mutual information‐based medical image registration using a hybrid optimisation technique , 2018, Electronics Letters.

[10]  Dittmar Böckler,et al.  Fusion Imaging to Support Endovascular Aneurysm Repair Using 3D-3D Registration , 2016, Journal of endovascular therapy : an official journal of the International Society of Endovascular Specialists.

[11]  Andy King,et al.  Optimising the Volgenant-Jonker algorithm for approximating graph edit distance , 2017, Pattern Recognit. Lett..

[12]  Vivek Aggarwal,et al.  Integrating Morphological Edge Detection and Mutual Information for Nonrigid Registration of Medical Images. , 2019, Current medical imaging reviews.

[13]  Yan Ke,et al.  PCA-SIFT: a more distinctive representation for local image descriptors , 2004, CVPR 2004.

[14]  Luc Van Gool,et al.  Speeded-Up Robust Features (SURF) , 2008, Comput. Vis. Image Underst..

[15]  Marcel Salathé,et al.  Using Deep Learning for Image-Based Plant Disease Detection , 2016, Front. Plant Sci..

[16]  Yong-Sheng Chen,et al.  A novel point cloud registration using 2D image features , 2017, EURASIP J. Adv. Signal Process..

[17]  Fred L. Bookstein,et al.  Principal Warps: Thin-Plate Splines and the Decomposition of Deformations , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  Robert B. Fisher,et al.  3D non-rigid registration using color: Color Coherent Point Drift , 2018, Comput. Vis. Image Underst..

[19]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[20]  余先川 Yu Xian-chuan,et al.  Review of remote sensing image registration techniques , 2013 .

[21]  David G. Lowe,et al.  Object recognition from local scale-invariant features , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[22]  Junjun Jiang,et al.  Robust Feature Matching for Remote Sensing Image Registration via Locally Linear Transforming , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[23]  Jean-Michel Morel,et al.  ASIFT: A New Framework for Fully Affine Invariant Image Comparison , 2009, SIAM J. Imaging Sci..

[24]  S. Arivazhagan,et al.  Detection of unhealthy region of plant leaves and classification of plant leaf diseases using texture features , 2013 .

[25]  R. R. Bhavani,et al.  Image registration for varicose ulcer classification using KNN classifier , 2017 .

[26]  Zheng Tian,et al.  Remote sensing image registration based on KICA-SIFT descriptors , 2010, 2010 Seventh International Conference on Fuzzy Systems and Knowledge Discovery.

[27]  Yongjun Zhang,et al.  Fine Registration for VHR Images Based on Superpixel Registration-Noise Estimation , 2018, IEEE Geoscience and Remote Sensing Letters.