论文信息 - Using Feedforward Neural Networks for Color Based Grape Detection in Field Images

Using Feedforward Neural Networks for Color Based Grape Detection in Field Images

Precise yield estimation is a relevant challenge for agriculture. In this paper we apply simple feedforward neural networks (FFNN) on the problem of color-based grape detection in field images, achieving an average classification rate of 93% near realtime. Our evaluation shows a detailed comparison between our FFNN approach and SVMs handled by up to date SVM implementations, revealing FFNN can slightly outperform SVMs regarding computation time while obtaining competitive results. Furthermore our results are not only competitive with state of the art results of pixelwise color-based classification in the application field of precision farming but show also new contributions by investigating the influence of using different color models and evaluating our classifiers on different lighting conditions and grape varieties.

Vincent Casser | Vincent Casser

[1] Paulus Insap Santosa,et al. Neural Network Application on Foliage Plant Identification , 2011, ArXiv.

[2] Yuxuan Wang,et al. A Leaf Recognition Algorithm for Plant Classification Using Probabilistic Neural Network , 2007, 2007 IEEE International Symposium on Signal Processing and Information Technology.

[3] Thorsten Joachims,et al. Cutting-plane training of structural SVMs , 2009, Machine Learning.

[4] Scott E. Fahlman,et al. An empirical study of learning speed in back-propagation networks , 1988 .

[5] Scarlett Liu,et al. Towards Automated Yield Estimation in Viticulture , 2013 .

[6] Sanjiv Singh,et al. Automated Visual Yield Estimation in Vineyards , 2014, J. Field Robotics.

[7] Abdesselam Bouzerdoum,et al. Skin segmentation using color pixel classification: analysis and comparison , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8] Ribana Roscher,et al. Automated Image Analysis Framework for the High-Throughput Determination of Grapevine Berry Sizes Using Conditional Random Fields , 2017, ArXiv.

[9] Martin A. Riedmiller,et al. RPROP - A Fast Adaptive Learning Algorithm , 1992 .

[10] Kazuhiro Nakano,et al. Application of neural networks to the color grading of apples , 1997 .