Deep Learning based Corn Kernel Classification

This paper presents a full pipeline to classify sample sets of corn kernels. The proposed approach follows a segmentation-classification scheme. The image segmentation is performed through a well known deep learning-based approach, the Mask R-CNN architecture, while the classification is performed through a novel-lightweight network specially designed for this task—good corn kernel, defective corn kernel and impurity categories are considered. As a second contribution, a carefully annotated multitouching corn kernel dataset has been generated. This dataset has been used for training the segmentation and the classification modules. Quantitative evaluations have been performed and comparisons with other approaches are provided showing improvements with the proposed pipeline.

[1]  Wei Li,et al.  Combining discriminant analysis and neural networks for corn variety identification , 2010 .

[2]  Rafael Rieder,et al.  Computer vision and artificial intelligence in precision agriculture for grain crops: A systematic review , 2018, Comput. Electron. Agric..

[3]  Min Huang,et al.  Maize seed classification using hyperspectral image coupled with multi-linear discriminant analysis , 2019 .

[4]  Zulay Niño,et al.  Efecto del Almacenamiento de Granos de Maíz (Zea mays) sobre la Calidad del Aceite Extraído , 2009 .

[5]  Hubert Konik,et al.  Automated visual grading of grain kernels by machine vision , 2015, International Conference on Quality Control by Artificial Vision.

[6]  JingTao Yao,et al.  Toward a Three-Way Image Classification Model: A Case Study on Corn Grain Images , 2019, 2019 IEEE International Symposium on Multimedia (ISM).

[7]  M Effendi,et al.  Corn quality identification using image processing with k-nearest neighbor classifier based on color and texture features , 2019, IOP Conference Series: Earth and Environmental Science.

[8]  Siddhartha S. Mehta,et al.  Deep Orange: Mask R-CNN based Orange Detection and Segmentation , 2019, IFAC-PapersOnLine.

[9]  Yang Yu,et al.  Fruit detection for strawberry harvesting robot in non-structural environment based on Mask-RCNN , 2019, Comput. Electron. Agric..

[10]  Josse De Baerdemaeker,et al.  Hyperspectral waveband selection for on-line measurement of grain cleanness , 2009 .

[11]  Jos B. T. M. Roerdink,et al.  The Watershed Transform: Definitions, Algorithms and Parallelization Strategies , 2000, Fundam. Informaticae.

[12]  Zhenbo Li,et al.  Multi-kernel dictionary learning for classifying maize varieties , 2018 .

[13]  Xiaoming Li,et al.  Corn Classification System based on Computer Vision , 2019, Symmetry.

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

[15]  Paul J. Williams,et al.  Near infrared hyperspectral imaging in quality and safety evaluation of cereals , 2018, Critical reviews in food science and nutrition.

[16]  Giorgia Foca,et al.  Automated quantification of defective maize kernels by means of Multivariate Image Analysis , 2018 .

[17]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Wei Wang,et al.  Extended-Maxima Transform Watershed Segmentation Algorithm for Touching Corn Kernels , 2013 .

[19]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[20]  Jeremiah W. Johnson,et al.  Adapting Mask-RCNN for Automatic Nucleus Segmentation , 2018, ArXiv.

[21]  Eleni Vrochidou,et al.  Machine Vision Systems in Precision Agriculture for Crop Farming , 2019, J. Imaging.

[22]  Yang Tao,et al.  Automatic inspection machine for maize kernels based on deep convolutional neural networks , 2019, Biosystems Engineering.

[23]  Ross B. Girshick,et al.  Mask R-CNN , 2017, 1703.06870.

[24]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.