Moldy Rice Detection Method Based on Near Infrared Spectroscopy Image Processing Technology

Feng-rui WEN; Hai-ou GUAN; Xiao-dan MA; Feng ZUO; Li-li QIAN

doi:10.3964/j.issn.1000-0593(2022)02-0428-06

Journals >Spectroscopy and Spectral Analysis >Volume 42 >Issue 2 >Page 428 > Article

Spectroscopy and Spectral Analysis
Vol. 42, Issue 2, 428 (2022)

Moldy Rice Detection Method Based on Near Infrared Spectroscopy Image Processing Technology

Feng-rui WEN^1、*, Hai-ou GUAN^{1、1; *;}, Xiao-dan MA^1、1;, Feng ZUO^{2、2; 3; *;}, and Li-li QIAN^{2、2; 3; 4;}

Author Affiliations

¹1. College of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163000, China

²2. College of Food Science, Heilongjiang Bayi Agricultural University, Daqing 163000, China

show less

DOI: 10.3964/j.issn.1000-0593(2022)02-0428-06 Cite this Article

Feng-rui WEN, Hai-ou GUAN, Xiao-dan MA, Feng ZUO, Li-li QIAN. Moldy Rice Detection Method Based on Near Infrared Spectroscopy Image Processing Technology[J]. Spectroscopy and Spectral Analysis, 2022, 42(2): 428 Copy Citation Text

show less

Fig. 1. Near-infrared images of rice with different degrees of mildew
(a): Healthy; (b): Mild mildew; (c) Moderate mildew

Download full size | View in the Article

Fig. 2. Distribution of texture characteristics of three mildew states of rice
(a): Healthy; (b): Mild mildew; (c): Moderate mildew

Download full size | View in the Article

Fig. 3. Comparison of rice reflectance distribution with different mildew degrees

Download full size | View in the Article

Fig. 4. Output of rice mildew detection model

Download full size | View in the Article

编号	特征值	计算公式	作用说明
(1)	均值	$M = \overset{L - 1}{\sum_{i = 0}} Z_{i} P (Z_{i})$	霉菌分布后籽粒表面明亮程度
(2)	标准差	$σ = \sqrt[]{\sum (Z_{i} {- M)}^{2} P (Z_{i})}$	图像整体像元值相对于均值的离散程度
(3)	平滑度	$S = 1 - \frac{1}{1 + σ^{2}}$	稻谷NIR图像的粗糙程度
(4)	三阶距	$δ_{3} = \overset{L - 1}{\sum_{i = 0}} (Z_{i} {- M)}^{3} P (Z_{i})$	稻谷NIR图像直方图的对称性
(5)	一致性	$H = \overset{L - 1}{\sum_{i = 0}} P^{2} (Z_{i})$	稻谷NIR图像的规则程度
(6)	信息熵	$e = - \overset{L - 1}{\sum_{i = 0}} P (Z_{i}) lo g_{2} P (Z_{i})$	稻谷NIR图像中反射值的随机性

Table 1. Calculated texture features of rice NIR image

Download Citation

Save the article for my favorites

Paper Information