文献类型：期刊文献

中文题名：基于激光散斑STCD算法的种子活力分类方法研究

英文题名：Research on seed viability classification method based on laser speckle STCD algorithm

作者：门森[1];李潇鑫[1];张君豪[1];刘玥[1];刘冰冰[1,2]

第一作者：门森

机构：[1]北京联合大学机器人学院,北京市100020;[2]北京市信息服务工程重点实验室,北京市100020

第一机构：北京联合大学机器人学院

年份：2026

卷号：7

期号：1

起止页码：75-85

中文期刊名：智能化农业装备学报(中英文)

外文期刊名：Journal of Intelligent Agricultural Mechanization

基金：国家自然科学基金(31770769);北京市教育委员会科研计划项目(KM201911417008)。

语种：中文

中文关键词：激光散斑;时空耦合差分;Faster R-CNN;种子活力;无损检测

外文关键词：laser speckle;spatiotemporal coupled difference;Faster R-CNN;seed vigor;non-destructive testing

摘要：为解决传统种子活力检测方法速度慢,耗时长,成本高的问题,提出了一种将激光散斑技术、STCD和深度学习模型相结合的种子活力无损检测方法,构建了“图像采集-特征处理-模型训练-活力判定”一体化种子活力检测系统。采用632.8 nm He-Ne激光器、CCD相机搭建成像系统,获取玉米种子激光散斑图像,经过图像预处理,将单个玉米种子区域提取出来,后采用二维小波变换进行散斑图像去噪,之后利用STCD进行图像数据处理,将STCD与图像差分改进策略深度融合,将玉米种子激光散斑图像的特征信息凸显出来,以前4帧加权平均为参考帧,引入3×3邻域局部方差加权机制与遗忘因子α=0.7构建累积差分图,最终得到1600张差分图像,活种子与失活种子各800张,按7:2:1划分数据集。模型方面,采用MobileNetV3为骨干网络,构建轻量化Faster R-CNN模型,调整参数,优化训练策略至验证集准确率稳定。后用最优训练模型进行玉米种子散斑图像目标检测、特征提取与深度学习,最终完成种子分类回归,实现对种子活力的高效准确检测。结果表明,种子激光散斑图像经STCD处理后特征信息明显,差分图像信噪比最高达35.8 dB,经过轻量化Faster R-CNN模型训练后,活种子分类准确率达92.25%、失活种子分类准确率达93.25%,模型ROC曲线中AUC值98%,F1分数为92.63%,假阳性率为4.75%,单张图像识别时间为0.0315 s,比原模型单张检测时间缩短了0.0493 s,检测速度比原模型提高2.5倍,收敛轮数25,比原模型减少一半以上,训练效率高。证明该方法可以实现种子活力的快速无损检测。
To overcome the challenges of slow speed,prolonged time consumption,and high costs in traditional seed vigor detection methods,we propose a non-destructive seed vigor detection approach that integrates laser speckle technology,a STCD algorithm,and a deep learning model,which facilitate the construction of a comprehensive seed vigor detection system encompassing“image acquisition,feature processing,model training,and vigor determination.”An imaging system was developed by utilizing a 632.8 nm He-Ne laser and a CCD camera to capture laser speckle images of pea seeds.After image preprocessing,the individual pea seed area was extracted.Then,two-dimensional wavelet transform was applied to denoise the speckle images.Subsequently,the STCD method was utilized for image data processing.The STCD method is deeply integrated with the image difference improvement strategy to highlight the feature information of the pea seed laser spot image.The weighted average of the first 4 frames is taken as the reference frame.The 3×3 neighborhood local variance weighting mechanism and the forgetting factorα=0.7 are introduced to construct the cumulative difference map.Finally,1600 difference images are obtained,with 800 viable seeds and 800 non-viable seeds.Divide the dataset in a ratio of 7:2:1.In terms of the model,MobileNetV3 was adopted as the backbone network to construct the lightweight Faster R-CNN model.The parameters were adjusted and the training strategy was optimized until the Accuracy of the validation set was stable.Subsequently,the optimally trained model was employed for target detection,feature extraction,and deep learning on the speckle images of maize seeds.Ultimately,seed classification and regression were accomplished,enabling efficient and accurate detection of seed viability.The results indicate that the laser speckle images of seeds processed by STCD exhibit distinct feature information,with a maximum signal-to-noise ratio of 35.8 dB for the differential images.After training with the lightweight Faster R-CNN model,the classification accuracy for viable seeds is 92.25%and the accuracy for non-viable seeds is 93.25%.The model achieves an AUC value of 98%in the ROC curve,an F1-score of 92.63%,a false positive rate(FPR)of 4.75%.The recognition time for a single image is 0.0315 seconds,which is 0.0493 seconds shorter than that of the original model.The detection speed is 2.5 times faster than the original model,and the convergence epoch is 25,reduced by more than half compared to the original model,demonstrating high training efficiency.These findings confirm that the proposed method enables rapid and non-destructive detection of seed viability.