机器视觉在木材缺陷检测领域应用研究进展

为了提高木材利用率,采用机器视觉来实现木材缺陷快速而稳定的检测,不仅可以克服人工检测效率低、劳动强度大、准确率低等弊端,而且对提高木材加工企业的智能化水平具有重要意义。文中概述了机器视觉检测技术及设备的国内外研究现状,介绍基于机器视觉检测木材缺陷所涉及的相关理论和算法研究,以及相关图像处理算法的优缺点;针对机器视觉应用在木材缺陷检测领域存在的不足,提出机器视觉木材表面检测应进一步向人工智能方向发展,以提高木材缺陷检测效率及准确性。     

基于深度学习的小浆果一体化智能采摘与分选研究进展

小浆果因具有较高的营养价值和保健作用深受人们喜爱,被国际市场誉为第3代水果。随着市场需求量的增加,林业小浆果的智能采摘与分选成为一种新的趋势。目前,由于现有技术自动化和智能化程度不高,导致大量人力、物力的投入。深度学习方法的飞速发展促进了检测技术的精确作业和检测精度的提高。文中概述了深度学习的发展进程,通过总结国内外关于深度学习方法在小浆果采摘与分选中的应用,提出深度学习方法在实际应用过程中存在的问题及解决措施,并展望深度学习方法在小浆果智能采摘与分选检测方面的发展趋势。     

A field experiment on stabilization of Cd in contaminated soils by surface-modified nano-silica (SMNS) and its phyto-availability to corn and wheat

Journal of Soils and Sediments - Cd-contaminated soil is a common environmental problem. Stabilization is an effective way of in situ remediation for Cd contamination in soils, but there is a lack...     

Organic phosphorus mineralization characteristics in sediments from the coastal salt marshes of a Chinese delta under simulated tidal cycles

Journal of Soils and Sediments - Sediment organic phosphorus (OP) mineralization plays an important role in phosphorus cycling in coastal salt marshes. However, information on OP mineralization...     

Nitrogen transformation mediated by nitrate-dependent iron oxidation in anoxic freshwater

Journal of Soils and Sediments - Fe(III) transformation to Fe(II) via the nitrate-dependent iron oxidation process, occurring in anoxic sediments, has an important role in the nitrogen cycle. In...     

Variation in soil bulk density and hydraulic conductivity within a quasi-circular vegetation patch and bare soil area

Journal of Soils and Sediments - Soil bulk density and hydraulic conductivity are two of many key factors for vegetation establishment and the other way around. Variations in them may cause patchy...     

Effect of carbon rate and type amended with ammonium or nitrate on nitrous oxide emissions in a strong ammonia oxidation soil

Journal of Soils and Sediments - The work aimed to (1) better understand how C rate and type affecting N2O emissions when combined application with different N forms in a strong ammonia oxidation...     

Effects of freeze-thaw cycles and initial soil moisture content on soil aggregate stability in natural grassland and Chinese pine forest on the Loess Plateau of China

Journal of Soils and Sediments - Ongoing global warming is decreasing the thickness of snow covers and increasing initial soil moisture content (SMC), which increase the number of freeze-thaw...     

光谱特征变量和BP神经网络构建油用牡丹种子含水率估算模型

为了进一步提高种子含水率的高光谱估算精度,该研究测定了156份油用牡丹种子的近红外吸收光谱及其对应的含水率值,分析了近红外吸收光谱、一阶微分光谱、水分吸收特征参数与含水率的相关关系,构建了基于特征波长吸收光谱、特征波长一阶微分光谱、水分特征吸收参数和BP神经网络的油用牡丹种子含水率估算模型,并对模型进行了验证;再结合一元线性回归（SLR,Single Linear Regression）、逐步多元线性回归（SMLR,Stepwise MultipleLinear Regression）、偏最小二乘回归（PLSR,Partial Least Squares Regression）模型与BP神经网络（BPNN,BP Neural Network）模型进行比较。结果表明：1）油用牡丹种子含水率的吸收光谱特征波长位于1 410、1 900、1990 nm,一阶微分光谱特征波长位于1 150、1 950、2 080 nm;2）以DF2080和AD2140为自变量建立的一元线性回归模型预测效果较优,在能够满足水分估算精度的情况下,是最优的选择方法。3）将优选的特征参数作为输入,实测含水率值作为输出,构建BP神经网络模型,其建模与验模R2分别为0.978和0.973,RMSE分别为0.22%和0.242%,而RPD值分别为6.478和5.889,与其他模型相比,BP神经网络模型的建模及预测精度均最高,是估算油用牡丹种子含水率的最优模型,其次为逐步多元线性回归模型。研究结果表明BP神经网络模型对种子含水率具有更好的预测能力,是估算油用牡丹种子含水率的有效方法。