Soybean Appearance Quality Discrimination Based on Visible Spectrogram

Fig. 2. Visible spectrograms of soybeans are classified into three grades according to their appearance quality. (a) Good grade; (b) medium grade; (c) inferior grade

Download full size

Fig. 3. Conversion of RGB color space images of raw soybeans collected from 3CCD charge-coupled imager to YCbCr color space images

Download full size

Fig. 4. Visible spectrograms of soybeans with multi-scale spatial gradient characteristics. (a) Good grade; (b) medium grade

Download full size

Table 1. Modeling and prediction accuracies of soybean discrimination using single-modal and low-rank sparse coupling representation features, respectively%
View table
Table 1. Modeling and prediction accuracies of soybean discrimination using single-modal and low-rank sparse coupling representation features, respectively%
Method Modeling accuracy Prediction accuracy
Gradient-based+K-means+LMNN 79.2 64.4
YCbCr-based+K-means+LMNN 86.5 70.8
LRS-based+Kernel K-means+LMNN 92.7 80.1

Tools

Get Citation

Copy Citation Text

Ping Lin, Jianqiang He, Zhiyong Zou, Yongming Chen. Soybean Appearance Quality Discrimination Based on Visible Spectrogram[J]. Acta Optica Sinica, 2019, 39(8): 0815002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites