基于高光谱技术的猪肉肌红蛋白含量无损检测

王立舒; 胡金耀; 房俊龙; 陈曦; 李闯

doi:10.11975/j.issn.1002-6819.2021.16.035

基于高光谱技术的猪肉肌红蛋白含量无损检测

东北农业大学电气与信息学院，哈尔滨 150030

基金项目: 黑龙江省教育厅科技课题（12521038）

计量
- 文章访问数: 534
- HTML全文浏览量: 7
- PDF下载量: 585
出版历程
- 收稿日期: 2020-03-17
- 修回日期: 2021-08-12
- 发布日期: 2021-08-14

Non-destructive detection of pork myoglobin content based on hyperspectral technology

School of Electricity and Information, Northeast Agricultural University, Harbin 150030

摘要

摘要: 为充分利用猪肉光谱与图像信息，实现猪肉肌红蛋白含量的在线检测，该研究提出一种基于深度学习模型的猪肉肌红蛋白含量无损检测方法。采用高光谱设备采集冷藏过程中猪肉高光谱图像，通过ENVI5.3选择图像感兴趣区域（Region Of Interest，ROI），分别提取ROI平均光谱信息与主成分图像信息。利用卷积自动编码器（Convolutional Auto Encoder，CAE）提取光谱与图像信息深度特征，分别建立光谱特征、图像特征及图-谱融合特征与肌红蛋白含量之间关系的卷积神经网络（Convolutional Neural Network，CNN）预测模型。其中基于融合深度特征CNN预测模型准确度较高，该模型对脱氧肌红蛋白（DeoMb）、氧合肌红蛋白（OxyMb）、高铁肌红蛋白（MetMb）含量预测集决定系数分别为0.964 5、0.973 2、0.958 5，预测集均方根误差RMSEP分别为0.015 8、0.226 6、0.381 6。为进一步验证图-谱融合特征与猪肉肌红蛋白存在对应关系，分别建立偏最小二乘回归（Partial Least Squares Regression，PLSR）、支持向量机回归（Support Vector Regression，SVR）预测模型。结果表明：CAE能充分提取图像与光谱特征；基于融合特征建立回归模型能提高肌红蛋白含量预测精度，相比于光谱信息与图像信息，以MetMb为例，其分别提高5.42%、16.12%。该检测方法为肉类质量在线检测提供参考，具有好的应用前景。
- 无损检测 /
- 光谱特征 /
- 高光谱图片 /
- 卷积神经网络 /
- 卷积自编码器
Abstract: Abstract: Hyperspectral imaging system can widely be expected to acquire a set of sample images within certain spectral bands in each pixel at the same time. In this study, rapid detection was proposed for the myoglobin content in pork samples using spectral images and deep learning. The pork was placed under the cold storage conditions at 4°C, where a total of 250 pork samples were settled at different times (0-5 d). A hyperspectral imager was used to collect the pork hyperspectral images (400 to 1 000 nm). ENVI5.3 software was also selected to determine the region of interest (ROI) in the hyperspectral images, thereby extracting the full-band average spectrum and principal component image of ROI. Subsequently, a Savitzky-Golay (SG) filter was used to denoise the spectral information for the curve smoothness and spectral resolution. A convolutional auto encoder (CAE) was utilized to extract spectral depth features. A prediction model was finally established for the content of deoxymyolglobin (DeoMb), oxymyoglobin (OxyMb), and metmyoglobin (MetMb) in the pork samples. The results showed that the determination coefficients of test datasets were 0.923 8, 0.920 3, and 0.909 2, and the root mean square errors (RMSE) were 0.033 4, 0.619 7, and 0.809 1, respectively. Furthermore, the image information of adjacent wavelengths was highly correlated against the image extraction and storage. Principal Component Analysis (PCA) was utilized to reduce the dimension of hyperspectral images for better storage and processing. As such, the images under all bands were linearly combined to form a principal component image in the ENVI5.3 software. The first three principal component images represented 90.62% of the original hyperspectral image, where the contribution rate of the first principal component was 88.50%, indicating the most information. Therefore, the first principal component image was selected for the subsequent image extraction. The first principal component image was unified to the size of 16×16 pixels, and then converted into a 768-dimensional column vector for the extraction of image depth features using a convolutional encoder. DeoM, OxyMb, and MetMb content prediction models were established using image depth features, in which the determination coefficients of test datasets were 0.772 1, 0.828 7, and 0.825 4, while the RMSE of prediction were 0.105 8, 1.302 7, and 1.566 7. The spectral and image features were fused at the data level, and then the fusion data was input into the CAE to extract the deep fusion features. The DeoMb, OxyMb, and MetMb content prediction models were also established using the fusion depth features. The determination coefficients of test datasets were 0.964 5, 0.973 2, and 0.958 5, while the RMSE of prediction were 0.015 8, 0.226 6, and 0.381 6. Obviously, the determination coefficients of the test dataset were improved, while the RMSE were reduced, compared with the individual image and spectrum information. Partial least square regression (PLSR) and support vector machine regression (SVR) prediction models were also established to further verify the relationship between the graph-spectrum fusion feature and pork myoglobin. It was found that the determination coefficients of the test dataset were greater than 0.85. Consequently, the convolutional autoencoder can be expected to extract the deep fusion features of image and spectral information. Moreover, the fusion features can better reflect the internal and external information of pork. The CNN regression model using the fusion features can also be used to improve the prediction accuracy. This finding can provide a new better way to detect the myoglobin content in pork using hyperspectral imaging.
- nondestructive detection /
- spectral feature /
- hyperspectral image /
- convolutional neural network /
- convolutional autoencoder

HTML全文

参考文献(30)

[1]	汤晓艳，赵小丽，徐学万. 大宗猪肉质量分级现状与实施对策建议[J]. 农产品质量与安全，2020(5)：36-40.Tang Xiaoyan, Zhao Xiaoli, Xu Xuewan. Current status of bulk pork quality classification and suggestions for implementation[J]. Quality and Safety of Agro-products, 2020(5):36-40. (in Chinese with English abstract) .
[2]	Neethling N E, Suman S P, Sigge G O, et al. Exogenous and endogenous factors influencing color of fresh meat from ungulates[J]. Meat and Muscle Biology, 2017, 1(1):253-257.
[3]	马超. 贮藏期内滩羊熟肉肌红蛋白含量变化的高光谱模型构建及空间分布[D]. 宁夏：宁夏大学，2019.Ma Chao. Hyperspectral Model and Spatial distribution of Myoglobin Content in Cooked Meat of Tan Sheep During Storage Period[D], Ningxia: Ningxia University, 2019. (in Chinese with English abstract) .
[4]	Yuan R, Liu G, He J, et al. Determination of metmyoglobin in cooked tan mutton using Vis/NIR hyperspectral imaging system[J]. Journal of Food Science, 2020, 85(22):1403-1410.
[5]	唐海涛，孟祥添，苏循新，等. 基于CARS算法的不同类型土壤有机质高光谱预测[J]. 农业工程学报，2021，37(2)：105-113.Tang Haitao, Meng Xiangtian, Su Xunxin, et al. Hyperspectral prediction on soil organic matter of different types using CARS algorithm[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(2): 105-113. (in Chinese with English abstract).
[6]	杨越超，赵英俊，秦凯，等. 黑土养分含量的航空高光谱遥感预测[J]. 农业工程学报，2019，35(20)：94-101.Yang Yuechao, Zhao Yingjun, Qin Kai, et al. Prediction of black soil nutrient content based on airborne hyperspectral remote sensing[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(20): 94－101. (in Chinese with English abstract).
[7]	Wang Y, Ma H B, Wang J Z, et al. Hyperspectral monitor of soil chromium contaminant based on deep learning network model in the Eastern Junggar coalfield[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2021, 257: 119739.
[8]	邵园园，王永贤，玄冠涛，等. 高光谱成像快速检测壳聚糖涂膜草莓可溶性固形物[J]. 农业工程学报，2019，35(18)：245-254.Shao Yuanyuan, Wang Yongxian, Xuan Guantao, et al. Rapid detection of soluble solids content in strawberry coated with chitosan based on hyperspectral imaging[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(18): 245-254. (in Chinese with English abstract).
[9]	袁瑞瑞，刘贵珊，何建国，等. 可见近红外高光谱成像对灵武长枣定量损伤等级判别[J]. 光谱学与光谱分析，2021，41(04)：1182-1187.Yuan Ruirui, Liu Guishan, He Jianguo, et al. Quantitative damage identification of lingwu long jujube based on visible near-infrared hyperspectral imaging. spectroscopy and spectral analysis, 2021, 41(04): 1182-1187. (in Chinese with English abstract)
[10]	Kucha C T, Liu L, Ngadi M, et al. Hyperspectral imaging and chemometrics as a non-invasive tool to discriminate and analyze iodine value of pork fat[J]. Food Control, 2021,127: 108145.
[11]	Wan G, Liu G, He J, et al. Feature wavelength selection and model development for rapid determination of myoglobin content in nitrite-cured mutton using hyperspectral imaging[J]. Journal of Food Engineering, 2020, 287:110090.
[12]	孙俊，靳海涛，芦兵，等. 基于高光谱图像及深度特征的大米蛋白质含量预测模型[J]. 农业工程学报，2019，35(15)：295-303.Sun Jun, Jin Haitao, Lu Bing, et al. Prediction model of rice protein content based on hyperspectral image and deep feature[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(15): 295-303. (in Chinese with English abstract).
[13]	王彩霞，王松磊，贺晓光，等. 高光谱图谱融合检测羊肉中饱和脂肪酸含量[J]. 光谱学与光谱分析，2020，40(02)：595-601.Wang Caixia, Wang Songlei, He Xiaoguang, et al. Detection of saturated fatty acid content in mutton by using the fusion of hyperspectral Spectrum and image information[J]. Spectroscopy and Spectral Analysis, 2020, 40(2): 595-601. (in Chinese with English abstract).
[14]	翁士状，唐佩佩，张雪艳，等. 高光谱成像的图谱特征与卷积神经网络的名优大米无损鉴别[J]. 光谱学与光谱分析，2020，40(9)：2826-2833.Wong Shizhuang, Tang Peipei, Zhang Xueyan, et al. Non-destructive identification method of famous rice based on image and spectral features of hyperspectral imaging with convolutional neural network[J]. Spectroscopy and Spectral Analysis，2020，40(9)：2826-2833.
[15]	孙俊，靳海涛，武小红，等. 基于低秩自动编码器及高光谱图像的茶叶品种鉴别[J]. 农业机械学报，2018，49(8)：316-323.Sun Jun, Jin Haitao, Wu Xiaohong, et al. Tea variety identification based on low-rank stacked auto-encoder and hyperspectral image[J]. Transaction of the Chinese Society for Agricultural Machinery，2018，49(8): 316-323.(in Chinese with English abstract).
[16]	李晴晴，侯瑞春，丁香乾. 基于改进堆叠自编码器的滚动轴承故障诊断[J].计算机工程与设计，2019，40(7)：2064-2070.Li Qingqing，Hou Ruichun，Ding Xiangqian. Roller bearing fault diagnosis based on improved stacked auto-encoder[J]. Computer Engineering and Desing，2019,40(7):2064-2070.
[17]	史杨，王儒敬，汪玉冰. 利用改进自动编码器光谱法预测土壤有机质[J]. 发光学报，2018，39(10)：1458-1465.Shi Yang，Wang Rujing，Wang Yubing. Prediction of soil organic matter by improved auto encoder based on Near-infrared spectroscopy[J]. Chinese Journal of Luminescence，2018，39(10)：1458-1465. (in Chinese with English abstract).
[18]	Feng J, Liu L, Cao X, et al. Marginal stacked autoencoder with adaptively-spatial regularization for hyperspectral image classification[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 6: 15224-15235.
[19]	田启川，王满丽. 深度学习算法研究进展[J]. 计算机工程与应用， 2019，55(22)：25-33.Tian Qichuan, Wang Manli. Research progress on deep learning algorithms[J]. Computer Engineering and Applications, 2019, 55(22): 25-33. (in Chinese with English abstract).
[20]	张思雨，张秋菊，李可. 采用机器视觉与自适应卷积神经网络检测花生仁品质[J]. 农业工程学报，2020，36(4)：269-277.Zhang Siyu, Zhang Qiuju, Li Ke. Detection of peanut kernel quality based on machine vision and adaptive convolution neural network[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(4): 269-277. (in Chinese with English abstract).
[21]	Acquarelli J, Van Laarhoven T, Gerretzen J, et al. Convolutional neural networks for vibrational spectroscopic data analysis[J]. Analytica Chimica Acta, 2017, 954: 22-31.
[22]	Krzywicki K. The determination of haem pigments in meat[J]. Meat Science, 1982, 7(1): 29-36.
[23]	姜微. 高光谱技术在马铃薯品种鉴别及品质无损检测中的应用研究[D]. 哈尔滨：东北农业大学，2017.Jang Wei. Study on Nondestructive Detection of Identificaiton Varieties and Quality Potato Using Hyperspectral Technology[D]. Harbin: Northeast Agricultural University，2017. (in Chinese with English abstract).
[24]	王春雷，陈婉芝，卢彩云，等. 基于高光谱成像的玉米收获后根茬行分割方法[J]. 农业机械学报，2020，51(S2)：421-426.Wang Chunlei, Chen Wanzhi, Lu Caiyun, et al. Segmentation method for maize stubble row based on hyperspectral Imaging[J]. Transaction of the Chinese Society for Agricultural Machinery，2020, 51(S2): 421-426. (in Chinese with English abstract).
[25]	Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-507.
[26]	Chen Y, Lin Z, Xing Z, et al. Deep learning-based classification of hyperspectral data[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2017, 7(6): 2094-2107.
[27]	曲建岭，余路，袁涛，等. 基于一维卷积神经网络的滚动轴承自适应故障诊断算法[J]. 仪器仪表学报，2018，39(7)：134-143.Qu Jianling, Yu Lu, Yuan Tao, et al. Adaptive fault diagnosis algorithm for rolling bearings based on one-dimensional convolutional neural network[J]. Chinese Journal of Scientific Instrument, 2018, 39(7): 134-143. (in Chinese with English abstract).
[28]	陈健，刘明，熊鹏，等. 基于卷积自编码神经网络的心电信号降噪[J]. 计算机工程与应用，2020，56(16)：148-155.Chen Jian, Liu Ming, Xiong Peng, et al. ECG signal denoising based on convolutional auto-encoder neural network[J]. Computer Engineering and Applications, 2020, 56（16）：148-155. (in Chinese with English abstract).
[29]	周琪. 基于自编码器与卷积神经网络的遥感对象分类[D].武汉：武汉理工大学，2018.Zhou Qi. Classification of Remote Sensing Objects Based on Autoencoder and Convolutional Neural Network[D].Wuhan: Wuhan University of Technology，2018(in Chinese with English abstract).
[30]	李小占，马本学，喻国威，等. 基于深度学习与图像处理的哈密瓜表面缺陷检测[J]. 农业工程学报，2021，37(1)：223-232.Li Xiaozhan，Ma Benxue, Yu Guowei, et al. Surface defect detection of Hami melon using deep learning and image processing[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(1): 223-232. (in Chinese with English abstract).

施引文献(8)

期刊类型引用(5)

1.	左杰文，彭彦昆，李荣娇，徐汇丽. 猪肉新鲜度特征波长筛选及无损智能检测装置优化. 农业工程学报. 2025(01): 308-316 . 本站查看
2.	刘忠艳，郝博麟，刘瑶，乔付. 一种高效的蛤仔重金属污染检测方法. 制造业自动化. 2023(07): 22-25 . 百度学术
3.	康孝存，沈广辉，徐剑宏，马桂珍，史建荣. 玉米中伏马毒素B污染高光谱快速检测模型研究. 中国粮油学报. 2023(08): 41-48 . 百度学术
4.	张硕，韩少云，熊黎剑，侯宇鑫，高薪，汤修映. 基于气敏传感器阵列特征优化的储粮害虫赤拟谷盗检测. 农业工程学报. 2022(10): 303-309 . 本站查看
5.	宋科，杨崇龙，石永宏，张伟，宋进，戈永慧，潘磊庆，屠康. 基于可见-近红外高光谱技术对鸡种蛋性别鉴定的研究. 食品安全质量检测学报. 2022(17): 5518-5525 . 百度学术