基于图像处理的玉米破损比例检测方法及其品质变化同步分析

doi:10.3864/j.issn.0578-1752.2026.09.011

摘要/Abstract

摘要：

【目的】针对传统玉米破损比例检测方法（如手动挑选称重法、电动筛选器法等）耗时费力且无法满足储运过程中品质实时评估的需求，本研究提出一种基于图像处理与深度学习的玉米破损比例快速检测方法，并同步分析破损比例对玉米长距离运输储藏后品质变化的影响，旨在为玉米质量智能监测、风险预警及储运条件优化提供技术支撑与理论依据。【方法】以500张智能手机拍摄的原始玉米粒图像为数据源，利用图像处理技术（图像校正、图像预处理、图像分割以及面积提取），通过3种深度学习模型（Resnet50、Densenet121和ViT）的训练与评估，建立玉米破损比例快速预测模型；并通过模拟玉米“北粮南运”水、陆运输路线的温湿度条件，系统探究不同破损比例玉米（0—12%）储运后的品质变化规律。【结果】ViT模型在玉米破损籽粒识别中表现最优，准确率和精确率均达99%，预测值和真实值的平均绝对误差仅为0.45%，决定系数R²高达0.978。随着破损比例从0增至12%，水路和陆路玉米的理化特性指标均显著上升（P<0.05）：水分含量分别增加1.384%和0.461%，脂肪酸值分别增加7.92和4.49 mg KOH/100 g，电导率分别升高6.72和4.66 μs·cm^-1，丙二醛（malondialdehyde，MDA）含量分别上升38.73%和19.22%；在营养品质方面，玉米蛋白质含量降低，淀粉含量呈先升后降再升的波动趋势，直链淀粉（amylose，AM）含量与支链淀粉（amylopectin，AP）含量比例失衡，糊化特性也发生显著改变。【结论】本研究为玉米破损比例的快速检测提供了可靠的技术手段，也为优化玉米储运条件、减缓品质劣变提供了科学依据。建议在“北粮南运”等长距离高湿度储运场景中将破损比例4%作为管控阈值，对超阈值批次实施优先分拣或保质处理，以减少粮食产后损失、保障我国粮食安全。

关键词: 玉米, 破损比例, 图像处理, 深度学习, 品质变化, 快速检测

Abstract:

【Objective】Traditional methods for detecting the corn damage ratio (e.g., manual sorting and weighing, and electric sieving) are time-consuming and labor-intensive, failing to meet the demand for real-time quality assessment during storage and transportation. To address this, this study proposed a rapid detection method for the corn damage ratio based on image processing and deep learning. Furthermore, the impact of the damage ratio on corn quality deterioration following long-distance transportation and storage was simultaneously analyzed. This research aimed to provide a technical support and a theoretical basis for intelligent corn quality monitoring, early risk warning, and the optimization of storage and transportation conditions.【Method】A total of 500 raw corn kernels images captured by smartphones were utilized as the dataset. A rapid prediction model for the corn damage ratio was developed using image processing techniques (image rectification, preprocessing, segmentation, and area extraction) combined with the training and evaluation of three deep learning models (Resnet50, Densenet121, and Vision Transformer [ViT]). Additionally, by simulating the temperature and humidity conditions of the waterway and land routes in China’s “North-to-South Grain Transportation” system, the quality variation of corn with different damage ratios (0-12%) following storage and transportation was systematically investigated.【Result】The ViT model demonstrated optimal performance in identifying damaged corn kernels, achieving both accuracy and precision rates of 99%. The mean absolute error between the predicted and actual values was merely 0.45%, with a coefficient of determination (R²) reaching 0.978. As the damage ratio increased from 0 to 12%, the physicochemical properties of corn transported via waterway and land routes showed significant increases (P<0.05): moisture content rose by 1.384% and 0.461%, fatty acid values increased by 7.92 mg KOH/100 g and 4.49 mg KOH/100 g, electrical conductivity elevated by 6.72 and 4.66 μs·cm^-1, and malondialdehyde (MDA) content increased by 38.73% and 19.22%, respectively. Regarding nutritional quality, the protein content decreased, while the starch content exhibited a fluctuating trend of an initial increase, followed by a decrease, and a subsequent rise. An imbalance emerged between the amylose (AM) and amylopectin (AP) content, accompanied by significant alterations in the pasting properties.【Conclusion】This study provided a reliable technical approach for the rapid detection of the corn damage ratio, offering a scientific basis for optimizing storage and transportation conditions to mitigate quality deterioration. Specifically, it was recommend establishing a 4% damage ratio as the critical control threshold in long-distance and high-humidity storage and transportation scenarios, such as China’s “North-to-South Grain Transportation” system. Batches exceeding this threshold should undergo priority sorting or quality preservation treatments.

Key words: corn, damage ratio, image processing, deep learning, quality deterioration, rapid testing

王志高, 李刘滨, 许颖, 鞠澄辉, 何荣. 基于图像处理的玉米破损比例检测方法及其品质变化同步分析[J]. 中国农业科学, 2026, 59(9): 1987-2001.

WANG ZhiGao, LI LiuBin, XU Ying, JU ChengHui, HE Rong. Image Processing-Based Detection Method for Corn Damage Ratio and Synchronous Analysis of Quality Variation[J]. Scientia Agricultura Sinica, 2026, 59(9): 1987-2001.

图/表 14

图1

图2

图3

图4

图5

表1

表2

图6

图7

图8

图9

表3

表4

表5

参考文献 40

[1]	ZHANG H W, DU B, JIANG S C, ZHU J Q, WU Q X. Potential assessment of selenium for improving nitrogen metabolism, yield and nitrogen use efficiency in wheat. Agronomy, 2023, 13(1): 110. doi: 10.3390/agronomy13010110
[2]	ZAHRA N, HAFEEZ M B, NAWAZ A, FAROOQ M. Rice production systems and grain quality. Journal of Cereal Science, 2022, 105: 103463. doi: 10.1016/j.jcs.2022.103463
[3]	FAN C L, WANG W J, CUI T, LIU Y, QIAO M M. Maize kernel broken rate prediction using machine vision and machine learning algorithms. Foods, 2024, 13(24): 4044. doi: 10.3390/foods13244044
[4]	FAN C L, ZHANG D X, YANG L, CUI T, HE X T, DONG J Q, ZHAO H H. Power consumption and performance of a maize thresher with automatic gap control based on feed rate monitoring. Biosystems Engineering, 2022, 216: 147-164. doi: 10.1016/j.biosystemseng.2022.02.015
[5]	FERRARETTO L F, SHAVER R D, LUCK B D. Silage review: Recent advances and future technologies for whole-plant and fractionated corn silage harvesting. Journal of Dairy Science, 2018, 101(5): 3937-3951. doi: S0022-0302(18)30319-9 pmid: 29685271
[6]	师建芳, 翟晓娜, 王小萌, 邵广, 赵慧凝, 谢奇珍. 东北地区粮食产后质量问题及保质运输技术探讨. 粮油食品科技, 2021, 29(2): 200-210.
	SHI J F, ZHAI X N, WANG X M, SHAO G, ZHAO H N, XIE Q Z. Discussion on quality defects after harvest and the maintaining quality transportation technology of grain in Northeast China. Science and Technology of Cereals, Oils and Foods, 2021, 29(2): 200-210. (in Chinese)
[7]	王昊, 祝玉华, 李智慧, 甄彤. 机器视觉在农作物种子检测中的研究进展. 计算机工程与应用, 2023, 59(22): 69-83. doi: 10.3778/j.issn.1002-8331.2303-0166
	WANG H, ZHU Y H, LI Z H, ZHEN T. Research progress of machine vision in crop seed inspection. Computer Engineering and Applications, 2023, 59(22): 69-83. (in Chinese) doi: 10.3778/j.issn.1002-8331.2303-0166
[8]	HIMMELBOE M, JØRGENSEN J R, GISLUM R, BOELT B. Seed identification using machine vision: Machine learning features and model performance. Computers and Electronics in Agriculture, 2025, 231: 109884. doi: 10.1016/j.compag.2024.109884
[9]	MAHIRAH J, YAMAMOTO K, MIYAMOTO M, KONDO N, OGAWA Y, SUZUKI T, HABARAGAMUWA H, AHMAD U. Monitoring harvested paddy during combine harvesting using a machine vision-Double lighting system. Engineering in Agriculture, Environment and Food, 2017, 10(2): 140-149. doi: 10.1016/j.eaef.2016.12.001
[10]	金诚谦, 刘士坤, 陈满, 杨腾祥, 徐金山. 采用改进U-Net网络的机收大豆质量在线检测. 农业工程学报, 2022, 38(16): 70-80.
	JIN C Q, LIU S K, CHEN M, YANG T X, XU J S. Online quality detection of machine-harvested soybean based on improved U-Net network. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(16): 70-80. (in Chinese)
[11]	杨亮, 王卓, 白晓平, 高雷, 扈伟斌. 玉米籽粒破碎率在线采样装置设计. 农机化研究, 2019, 41(6): 121-124, 129.
	YANG L, WANG Z, BAI X P, GAO L, HU W B. Design of on-line sampling device for maize kernel broken rate. Journal of Agricultural Mechanization Research, 2019, 41(6): 121-124, 129. (in Chinese)
[12]	陈进, 张帅, 李耀明, 朱林军, 夏慧, 朱亚辉. 联合收获机水稻破碎籽粒及杂质在线识别方法. 中国农机化学报, 2021, 42(6): 137-144. doi: 10.13733/j.jcam.issn.2095-5553.2021.06.22
	CHEN J, ZHANG S, LI Y M, ZHU L J, XIA H, ZHU Y H. Research on online identification system of rice broken impurities in combine harvester. Journal of Chinese Agricultural Mechanization, 2021, 42(6): 137-144. (in Chinese)
[13]	胡东红, 汪浩, 艾君, 张玲, 张胜兰. 两种图像校正算法在实际应用中的比较. 计算机工程与应用, 2009, 45(13): 191-193, 226.
	HU D H, WANG H, AI J, ZHANG L, ZHANG S L. Comparisons of two kinds of image rectification algorithms. Computer Engineering and Applications, 2009, 45(13): 191-193, 226. (in Chinese)
[14]	ZONG Z, LIU G, ZHAO S. Real-time localization approach for maize cores at seedling stage based on machine vision. Agronomy, 2020, 10(4): 470. doi: 10.3390/agronomy10040470
[15]	GAO J M, LIU C Z, HAN J Y, LU Q L, WANG H X, ZHANG J H, BAI X G, LUO J K. Identification method of wheat cultivars by using a convolutional neural network combined with images of multiple growth periods of wheat. Symmetry, 2021, 13(11): 2012. doi: 10.3390/sym13112012
[16]	金哲毅, 杨晓玲. 基于深度残差网络的新冠肺炎CT影像分类系统. 信息技术与信息化, 2021(10): 114-116.
	JIN Z Y, YANG X L. COVID-19 CT image classification system based on deep residual network. Information Technology and Informatization, 2021(10): 114-116. (in Chinese)
[17]	吉训生, 江昆, 谢捷. 基于多维神经网络深度特征融合的鸟鸣识别算法. 信号处理, 2022, 38(4): 844-853.
	JI X S, JIANG K, XIE J. Deep feature fusion of multi-dimensional neural network for bird call recognition. Journal of Signal Processing, 2022, 38(4): 844-853. (in Chinese)
[18]	涂雪滢, 张佳鹏, 钱程, 刘世晶. 融合ViT和通道注意力的水稻病害识别技术研究. 农业与技术, 2025, 45(7): 69-74.
	TU X Y, ZHANG J P, QIAN C, LIU S J. Research on rice disease identification technology combining ViT and channel attention. Agriculture and Technology, 2025, 45(7): 69-74. (in Chinese)
[19]	向莹莹, 李浩卓, 张婷婷, 王建华, 孙群. 电导率法早期检测玉米和小麦种子活力. 中国农业大学学报, 2020, 25(6): 12-19.
	XIANG Y Y, LI H Z, ZHANG T T, WANG J H, SUN Q. Study on early detection of seed vigor of maize and wheat seeds by conductivity method. Journal of China Agricultural University, 2020, 25(6): 12-19. (in Chinese)
[20]	GHIMIRE A, KIM S H, CHO A, JANG N, AHN S, ISLAM M S, MANSOOR S, CHUNG Y S, KIM Y. Automatic evaluation of soybean seed traits using RGB image data and a Python algorithm. Plants, 2023, 12(17): 3078. doi: 10.3390/plants12173078
[21]	杜岳峰, 侯思余, 李国润. 基于深度学习的玉米籽粒含杂检测方法研究. 农机化研究, 2025, 47(9): 1-8.
	DU Y F, HOU S Y, LI G R. Detection method of maize kernel impurity based on deep learning. Journal of Agricultural Mechanization Research, 2025, 47(9): 1-8. (in Chinese)
[22]	WANG Y G, LIANG H, WEN C C, ZHAO S R. Ts-vit: Feature- enhanced transformer via token selection for fine-grained image recognition. Signal, Image and Video Processing, 2024, 19(2): 127. doi: 10.1007/s11760-024-03693-1
[23]	TONG W, CHEN W T, HAN W, LI X J, WANG L Z. Channel- attention-based DenseNet network for remote sensing image scene classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 4121-4132. doi: 10.1109/JSTARS.4609443
[24]	DU X W, SI L Q, LI P F, YUN Z H. A method for detecting the quality of cotton seeds based on an improved ResNet50 model. PLoS ONE, 2023, 18(2): e0273057. doi: 10.1371/journal.pone.0273057
[25]	李志杰, 胡经营, 张静, 朱晨辉, 李世龙, 刘冰杰, 张红梅. 基于改进Yolov8的玉米籽粒破碎率检测装置设计与试验. 河南农业大学学报, 2025, 59(5): 893-903.
	LI Z J, HU J Y, ZHANG J, ZHU C H, LI S L, LIU B J, ZHANG H M. Design and test of corn kernel breakage rate detection device based on improved Yolov8. Journal of Henan Agricultural University, 2025, 59(5): 893-903. (in Chinese)
[26]	XU Y Z, MA P G, NIU Z P, LI B B, LV Y Y, WEI S, HU Y S. Effects of artificial aging on physiological quality and cell ultrastructure of maize (Zea mays L.). Cereal Research Communications, 2023, 51(3): 615-626. doi: 10.1007/s42976-022-00328-4
[27]	ZHANG J M, XING Z Z, GU F X, WANG Y L, WANG T B, CHEN J Y. Changes in the microflora on the seed surface and seed vigor of maize (Zea mays) under different conditions. PLoS ONE, 2024, 19(11): e0311258. doi: 10.1371/journal.pone.0311258
[28]	叶佳琳, 杜京霖, 吴光玥, 胡家涵, 王澳希, 吕昭颖, 胡浩. 温度及水分含量对二氧化碳气调储藏稻谷品质的影响. 中国稻米, 2024, 30(6): 60-65. doi: 10.3969/j.issn.1006-8082.2024.06.010
	YE J L, DU J L, WU G Y, HU J H, WANG A X, LÜ Z Y, HU H. Effects of storage temperature and rice moisture content on quality of rice in carbon dioxide atmosphere storage. China Rice, 2024, 30(6): 60-65. (in Chinese) doi: 10.3969/j.issn.1006-8082.2024.06.010
[29]	吴新连, 王大枚, 罗柏流, 张荣森, 李国坚. 不同破损粒含量玉米陈化试验. 粮食储藏, 2009, 38(3): 43-45.
	WU X L, WANG D M, LUO B L, ZHANG R S, LI G J. Effect of breakage content on corn aging. Grain Storage, 2009, 38(3): 43-45. (in Chinese)
[30]	陈霞, 王习政, 苏文娟, 谢荟清, 陈尚钘, 刘娟. 嫁接方法与套袋处理对山苍子生长、光合特性及叶绿素荧光特征的影响. 西北农林科技大学学报(自然科学版), 2024, 52(3): 51-61, 72.
	CHEN X, WANG X Z, SU W J, XIE H Q, CHEN S X, LIU J. Effects of grafting methods and bagging on growth, photosynthetic and chlorophyll fluorescence characteristics of Litsea cubeba. Journal of Northwest A & F University (Natural Science Edition), 2024, 52(3): 51-61, 72. (in Chinese)
[31]	张永辉, 何思雨, 王倩, 孙超, 彭亚平, 艾鹏飞. 谷外糙米含量在储藏过程中对稻谷品质的影响. 现代食品, 2022, 28(13): 153-155, 160.
	ZHANG Y H, HE S Y, WANG Q, SUN C, PENG Y P, AI P F. Effects of brown rice content outside the grain on the quality of rice during storage. Modern Food, 2022, 28(13): 153-155, 160. (in Chinese)
[32]	李兴岩, 赵薇, 薛敏, 李爱光, 何宛宸, 段慧敏, 刘霞. 偏高水分玉米贮藏技术及品质控制研究进展. 农产品加工(下半月), 2022(9): 101-103, 107.
	LI X Y, ZHAO W, XUE M, LI A G, HE W C, DUAN H M, LIU X. Advances in storage technology and quality control of maize with high moisture content. Agricultural Roducts Processing, 2022(9): 101-103, 107. (in Chinese)
[33]	张明丹. 不同储藏温湿度对玉米品质特性的影响[D]. 郑州: 河南工业大学, 2021.
	ZHANG M D. Effects of different storage temperature and humidity on quality characteristics of maize[D]. Zhengzhou: Henan University of Technology, 2021. (in Chinese)
[34]	张玉荣, 马记红, 周显青, 伦利芳, 褚洪强, 付玲. 充二氧化碳气调解除后大米储藏品质变化研究. 河南工业大学学报(自然科学版), 2013, 34(2): 29-33.
	ZHANG Y R, MA J H, ZHOU X Q, LUN L F, CHU H Q, FU L. Change on storage quality of rice after controlled atmosphere storage with CO₂. Journal of Henan University of Technology (Natural Science Edition), 2013, 34(2): 29-33. (in Chinese)
[35]	曾德斌, 王军民, 吴向阳, 韩秀秀, 龙永红, 李旭春, 张英, 潘典进. 青贮玉米不同品种和生育时期植株体的营养品质分析. 湖北农业科学, 2019, 58(5): 85-87, 95.
	ZENG D B, WANG J M, WU X Y, HAN X X, LONG Y H, LI X C, ZHANG Y, PAN D J. Analysis on the nutritional quality of plant body of silage maize with different varieties and different growth stages. Hubei Agricultural Sciences, 2019, 58(5): 85-87, 95. (in Chinese)
[36]	LIU C S, ZHOU Y, CHEN G X, ZHENG D Q, YUE L F. Compression and fungal heat production in maize bulk considering kernel breakage. Applied Sciences, 2022, 12(10): 4870. doi: 10.3390/app12104870
[37]	林颖雪. 大米霉变过程中营养成分变化及霉变预警研究[D]. 天津: 天津科技大学, 2020.
	LIN Y X. Study on changes of nutrient components and mildew early warning during rice mildew[D]. Tianjin: Tianjin University of Science & Technology, 2020. (in Chinese)
[38]	詹锦玲, 贾梦, 蔡奕盛, 杨天一, 麻荣荣, 田耀旗. 细胞壁破损对马铃薯细胞胞内淀粉理化特性和消化特性的影响. 食品工业科技, 2025, 46(18): 194-200.
	ZHAN J L, JIA M, CAI Y S, YANG T Y, MA R R, TIAN Y Q. Effects of cell wall damage on physicochemical and digestibility properties of intracellular starch in potato cells. Science and Technology of Food Industry, 2025, 46(18): 194-200. (in Chinese)
[39]	曹俊, 蒋伟鑫, 刘欣, 陈文若, 戴炳业, 董文, 陈银基. 温度动态变化对不同水分含量稻谷主要品质变化的影响. 食品科学, 2016, 37(17): 76-83. doi: 10.7506/spkx1002-6630-201617013
	CAO J, JIANG W X, LIU X, CHEN W R, DAI B Y, DONG W, CHEN Y J. Effects of dynamic temperature changes on main quality attributes of paddy rice with different moisture contents. Food Science, 2016, 37(17): 76-83. (in Chinese) doi: 10.7506/spkx1002-6630-201617013
[40]	罗财伟, 王茂飞, 宁冬, 王海彦, 袁建敏. 不同类型和含水量玉米储存期间理化性质及糊化特性变化研究. 中国畜牧杂志, 2023, 59(2): 207-214.
	LUO C W, WANG M F, NING D, WANG H Y, YUAN J M. Study on the changes of physicochemical properties and gelatinization characteristics of different types of corn with different moisture content during storage. Chinese Journal of Animal Science, 2023, 59(2): 207-214. (in Chinese)

模型 Model	准确率 Accuracy (%)	精确率 Precision (%)	召回率 Recall (%)	F₁值 F₁-Score (%)
Resnet50	89.67	89.40	90.00	90.20
Densenet121	95.33	94.74	96.00	95.37
ViT	99.00	99.33	98.67	98.99

编号 Number	真实值 Actual value			预测值 Predictive value			绝对误差Absolute error (%)	平均绝对误差 Mean absolute error (%)
编号 Number	完整质量 Total quality	破损质量 Damage quality	破损比例 Damage rate (%)	完整质量 Total quality	破损质量 Damage quality	破损比例 Damage rate (%)	绝对误差Absolute error (%)	平均绝对误差 Mean absolute error (%)
1	20.107	0.446	2.17	19.978	0.495	2.42	0.25	0.45
2	22.485	1.047	4.45	22.133	1.104	4.75	0.30
3	19.416	1.209	5.86	19.031	1.301	6.40	0.54
4	21.728	1.838	7.80	20.969	1.945	8.49	0.69
5	22.318	2.444	9.87	21.552	2.555	10.60	0.73
6	25.714	3.427	11.76	25.526	3.333	11.55	0.21

路线 Route	破损比例 Damage rate (%)	峰值黏度 Peak viscosity (cP)	最低黏度 Minimum viscosity (cP)	最终黏度 Final viscosity (cP)	衰减值 Attenuation value (cP)	回生值 Revival value (cP)
水路 Waterway	0	2019±42ab	1532±30ab	2966±65ab	487±23ab	1434±39a
	2	2090±21a	1565±10a	3055±44a	524±12a	1489±39a
	4	1961±46b	1499±34ab	2894±47ab	462±13b	1395±13ab
	6	1870±8bc	1442±16b	2744±15b	427±12b	1302±22b
	8	1768±19c	1389±26bc	2594±50bc	379±9c	1205±26bc
	10	1719±35c	1338±30bc	2569±94bc	380±17c	1231±64bc
	12	1686±47c	1307±39c	2444±72c	379±15c	1137±33c
陆路 Land route	0	1637±38b	1199±38a	2340±37a	438±1b	1141±31a
	2	1655±37ab	1217±44a	2350±82a	438±17b	1133±54a
	4	1606±60b	1229±67a	2419±65a	377±13c	1190±16a
	6	1668±15ab	1211±10a	2357±93a	456±5b	1145±84a
	8	1737±4ab	1250±5a	2481±24a	487±8ab	1231±21a
	10	1672±20ab	1236±12a	2469±10a	436±9b	1233±2a
	12	1762±72a	1234±47a	2465±58a	528±26a	1231±27a

指标 Indicators	破损比例 Damage rate	水分含量 Moisture content	脂肪酸值 Acid value	电导率 Conductivity	丙二醛 MDA	蛋白含量 Protein content	淀粉含量 Starch content	直链淀粉 Amylose	支链淀粉 AP
破损比例 Damage rate	1
水分含量 Moisture content	0.933^**	1
脂肪酸值 Acid value	0.977^**	0.897^**	1
电导率 Conductivity	0.991^**	0.920^**	0.966^**	1
丙二醛 MDA	0.967^**	0.888^**	0.914^**	0.937^**	1
蛋白含量 Protein content	-0.946^**	-0.898^**	-0.939^**	-0.944^**	-0.904^**	1
淀粉含量 Starch content	0.352	0.337	0.466	0.406	0.324	-0.498	1
直链淀粉 Amylose	0.865^*	0.978^**	0.849^*	0.867^*	0.807^*	-0.871^*	0.372	1
支链淀粉 Amylopectin	-0.868^*	0.978^**	-0.852^*	-0.870^*	-0.809^*	0.871^*	-0.366	-1^**	1

指标 Indicator	破损比例Damage rate	水分含量Moisture content	脂肪酸值Acid value	电导率Conductivity	丙二醛MDA	蛋白含量Protein content	淀粉含量Starch content	直链淀粉Amylose	支链淀粉Amylopectin
破损比例 Damage rate	1
水分含量 Moisture content	0.824^*	1
脂肪酸值 Acid value	0.983^**	0.837^*	1
电导率 Conductivity	0.948^**	0.693	0.908^**	1
丙二醛 MDA	0.848^*	0.886^**	0.890^**	0.73	1
蛋白含量 Protein content	-0.391	-0.358	-0.371^*	-0.359^*	-0.904^**	1
淀粉含量 Starch content	-0.019	0.189	0.059	0.146	0.324	0.335	1
直链淀粉 Amylose	-0.756^*	-0.483	-0.690	-0.451	0.807^*	0.288	0.612	1
支链淀粉 Amylopectin	0.765^*	0.493	0.703	0.459	-0.809^*	-0.278	-0.602	0.99^**	1