In this paper, the convolution neural network GoogLeNet is applied to sprouting potato recognition. In order to meet the needs of Screening for sprouting potatoes, image recognition technology is required. The traditional identification technology must describe the features of the objects specifically, but it is difficult to describe the features of complex potato shoots. Therefore, the deep neural network is used to identify the sprouting potatoes. In addition, the overall network model is applied on embedded devices for reducing the hardware expense. And the model on the embedded devices should have less training parameters and stronger expression capabilities than the model on PC. Therefore, the GooLeNet model is used to classify pictures. The research results show that the recognition rate of sprouting potatoes reached 80%, and the GooLeNet model has a great potential for sprouting potato recognition.
[1]
Ching Y. Suen,et al.
A trainable feature extractor for handwritten digit recognition
,
2007,
Pattern Recognit..
[2]
Geoffrey E. Hinton,et al.
ImageNet classification with deep convolutional neural networks
,
2012,
Commun. ACM.
[3]
Yoshua Bengio,et al.
Gradient-based learning applied to document recognition
,
1998,
Proc. IEEE.
[4]
Andrew Zisserman,et al.
Very Deep Convolutional Networks for Large-Scale Image Recognition
,
2014,
ICLR.
[5]
Abdesselam Bouzerdoum,et al.
An eye feature detector based on convolutional neural network
,
2005,
Proceedings of the Eighth International Symposium on Signal Processing and Its Applications, 2005..