基于YOLOv7的木材缺陷检测模型Wood-Net的研究

为改善利用人工方式识别木材缺陷存在的效率低、人工成本高的问题,同时实现在木材加工过程中使用新兴方式对不同的缺陷进行快速准确检测以提高木材利用率,针对现有的目标检测网络在木材缺陷检测方面存在诸如检测精度低、报错率高以及识别种类少等局限,设计了用于木材缺陷检测的深度学习网络Wood-Net。Wood-Net将注意力机制ECA(efficient channel attention module)引入YOLOv7的主干网络,以便更好地区分木材缺陷之间的细微差别;将ECA与Res2Net结合后形成ECA-Res2Net模块,ECA-Res2Net模块克服了单纯的Res2Net跨通道交流能力不足的问题,增强了网络对更细粒度特征的提取能力;将ECA-Res2Net模块与SPPCSPC(spatial pyramid pooling and channel spatial pyramid convolution)并联形成ResSPPCSPC模块,增加了描述图像本身特征数量的能力,由此构成新方法Wood-Net。本研究将准确度、召回值、mAP@0.5以及mAP@0.5∶mAP@0.95 4个数值作为系统性能的评价指标。利用自建数据集训练Wood-Net,得到试验数据。试验结果表明:Wood-Net模型比基准模型YOLOv7在木材优选上精确率提高了4.52%,mAP@0.5∶mAP@0.95提高了6.62%;比基准模型YOLOv5s在木材优选上精确率提高了6.79%,mAP@0.5∶mAP@0.95提高了5.67%。ECA注意力机制能够有效提升E-ELAN的通道间信息交互能力;Res2Net模块具有很强的细粒度特征提取能力,在网络中引入Res2Net模块后,网络各项性能指标收敛速度快,在Res2Net中加入ECA后能够使单纯的Res2Net考虑多通道特征之间的关系,完成信息融合,提高检测性能。

Wood灯检查在白癜风治疗期间不同复色模式观察研究中的应用

目的:通过观察Wood灯下白癜风患者治疗过程中皮损的复色模式,探究不同复色模式的相关影响因素,分析评估不同复色模式间的疗效差异。方法:收集选取2021年8月-2022年8月笔者医院就诊的90例(170块),采集患者基本信息及病情相关指标。治疗后,观察Wood灯下皮损复色模式,将复色模式与收集指标进行统计分析,筛选出差异有统计学意义的指标后,采用Logistic回归分析阳性指标与复色模式的相关性。并将Wood灯下观察的皮损面积作为疗效判定标准,分析对比不同复色模式间的疗效差异。结果:One-way ANOVA分析、χ^(2)检验显示白癜风疾病分期、皮损部位、治疗方式与皮损复色模式有关(P<0.05)。Logistic回归分析显示:治疗方式与复色模式存在相关性(P<0.05)。不同复色模式达到有效复色面积的时间比较,差异有统计学意义(H=8.04,P<0.05)。结论:通过观察Wood灯下白癜风患者治疗过程中皮损的复色模式发现,治疗方式可能是白癜风皮损复色模式的决定因素;不同复色模式之间的复色速度存在差异,可作为临床医师调整治疗方案的参考依据。

Physicochemical and Thermal Properties of Some Key Tropical Sawdust Woods for Energy Production

In South Saharan countries, 85% of the population uses biomass as a primary energy source. Cameroon presents one of the highest biomass energy and sawmills produce important sawdust resources which are not used and are burnt in piles leading to significant air toxic emissions. Therefore, we have to valorize industrially these available sawdusts. This study focuses on the physicochemical and thermochemical analysis of Ayous, Sapelli and Tali sawdust. The proximate and ultimate analysis, particle size, structural composition, as well as heavy metal content and calorific value were determined. In addition, the thermogravimetric mass losses were also estimated. The results showed that high water contents (24% - 41%) were recorded in the raw sawdust, and the thermal treatment reduced these contents from 78% to values in the range of 4% - 9%. The values for ash and volatile matter content were respectively between 0.25% - 0.74% and 68% - 76%. The LHV is higher in Ayous (17.5 MJ/kg) and Sapelli (16.8 MJ/kg) than that of Tali (15.7 MJ/kg). The concentration of heavy metals is very low in each species. Extractives are more present in Tali (16.06%) than in the other sawdusts. Pyrolysis of sawdust shows the typical decomposition of hemicellulose (270˚C - 325˚C), cellulose (325˚C - 400˚C) and lignin (200˚C - 550˚C) with a maximum loss of 75% at 370˚C and the melting point is 320˚C. The results of the sawdust parameters determined in the paper can be valorized to reduce pollutants emissions by developing the efficiency and effectiveness of biomass energy processes and promoting the use of biomass as a sustainable alternative to traditional fossil fuels.

Polyurethane Hybrid-Based Wood Adhesive: Review

Based on commercially available polyvinyl alcohol (PVA) stabilised polyvinyl acetate (PVAc), emulsion adhesives are neither heat nor moisture-resistant and show weak strength at high relative humidity and high temperatures. Pre- or post-crosslinking is another method used to manufacture a conventional vinyl-based homopolymers or copolymers system with improved water resistance. Vinyl neodecanoate (VeoVa), N-methylolacrylamide (NMA), Methacrylamide, methacrylic acid (MAA), and other self-crosslinking comonomers are typically inserted to produce highly water-resistant vinyl based homopolymers or copolymers. Additionally, organic crosslinkers like glyoxal, glutaraldehyde, citric acid, tartaric acid, and the like, as well as inorganic crosslinkers like acidic metal salts like aluminium chloride, aluminium nitrate, boric acid, and the like, can be used to prepare the highly water-resistant vinyl based homopolymers or copolymers. It is also possible to combine the self-crosslinking comonomers with the organic crosslinkers. Recently, a different hybrid chemistry has been developed that improves lap shear strength, has outstanding water resistance, good durability, and doesn’t require any additional crosslinker agents. Two distinct polymers were combined to develop hybrid polymers. They usually involve mixing an organic polymer with a polymer. There are many capping agents that are used for polyurethanes to produce acrylics that are capped with polyurethane and used as an oligomer in PVAc wood glue. Here, in this paper, we reviewed the different hybrid chemistry based on polyurethane chemistry for wood bonding applications.

Physico-Chemical Characterization of Wood Ash for Agronomic Purposes

The oil refinery of the food products company SOPAL, located in the port area of TOGO, produces an average of 200 kg of eucalyptus wood ash per day. This ash is stored in large quantities within the premises of the refinery, without any disposal solution. The aim of this study is to identify the physico-chemical properties of SOPAL’s wood ash for agronomic valorization. Physico-chemical analyses reveal the presence of major and secondary elements such as potassium (15.71 g/kg DM), phosphorus (27.90 g/kg DM), calcium (9.76 g/kg DM), magnesium (0.03 g/kg DM), sodium (1.24 g/kg DM), iron (1.49 g/kg DM) and manganese (7.82 g/kg DM). The pH is alkaline (12.97) with fairly high conductivity (24.10 mS/Cm). Heavy metals identified are cobalt (0.19 mg/kg DM), cadmium (0.10 mg/kg DM), lead (1.14 mg/kg DM), copper (19.77 mg/kg DM) and nickel (12.82 mg/kg DM). The results show that the ash contains considerable amount of major and secondary elements for agronomic uses.

马尔康6.0级震群地震藏族民居震害调查与分析

为掌握马尔康6.0级震群地震中藏族民居的震害情况,对地震灾区进行广泛的调查,总结藏族民居的建筑特点与结构特点,分析石木结构、木结构和生土结构的主要震害特征和破坏原因,并对不同结构类型的建筑震害数量和地质灾害隐患点数量进行统计和分析。研究结果表明:此次地震中,石木结构震害较严重,主要破坏特征为整体倒塌、局部倒塌、墙体破坏、梁柱破坏和屋盖破坏。木结构震害较轻,主要破坏特征为围护墙倒塌、开裂以及屋面溜瓦。由于距离震中较远,生土结构的震害最轻,主要破坏特征为墙体开裂。石木结构和木结构倒塌破坏占比分别为0.9%和0.0%,严重损坏占比分别为19.0%和17.2%。木结构多为非结构构件损坏,抗震能力强于石木结构。同时,地震诱发山体落石、山体滑坡,导致建筑被掩埋、道路中断、水利和通讯设施受损,地质灾害隐患点数量为663处。针对不同藏族民居的震害特征和破坏统计,建议在保留藏族民居建筑特色的前提下,发展受力机理清晰的结构体系,加强藏族民居的抗震措施,并重视该地区的震群型地震效应。根据次生灾害的破坏情况和统计,建议灾后重建中,藏族民居应合理选址,加强地质灾害的监测和预警,并完善防治次生灾害设施。

采用气相色谱-质谱指纹图谱鉴定铁木豆属木材

采用气相色谱-质谱指纹图谱分别鉴定平萼铁木豆、葱叶状铁木豆和马达加斯加铁木豆属木材。以二氯甲烷提取样品中的特定化合物,采用气相色谱-质谱法分析提取物,建立目标物色谱图。平萼铁木豆在保留时间25.1 min的最强吸收峰和22.3、22.6及23.4 min的次强吸收峰可作为平萼铁木豆的特征图谱,与其他两种铁木豆相区分。葱叶状铁木豆和马达加斯加铁木豆的最强吸收峰色谱保留时间均为16.3 min,葱叶状铁木豆的次强吸收峰色谱保留时间为18.5 min,另外还在色谱保留时间分别为19.4、19.8、24.3 min处有吸收峰,而马达加斯加铁木豆除在色谱保留时间17.7 min处有次强吸收峰外,还在色谱保留时间为23.4 min时有吸收峰。气相色谱-质谱法可用于平萼铁木豆、葱叶状铁木豆和马达加斯加铁木豆三类木材的鉴别。

APP/ZB协效阻燃发泡木塑复合材料性能研究

为了拓宽发泡木塑复合材料(FWPC)的功能性,以沙柳木粉、高密度聚乙烯(PE-HD)为主要原料,纳米炭黑(Nano-CB)为导电填料,聚磷酸铵(APP)和硼酸锌(ZB)作为阻燃剂和抑烟成分进行复配,采用发泡工艺和模压法制备阻燃抗静电型发泡木塑复合材料。研究APP/ZB协效阻燃剂质量比对FWPC力学性能、阻燃抑烟性能及热稳定性能的影响。结果表明:当APP/ZB总加入量为20%,质量比为4∶1时,FWPC的各项性能相对较优,静曲强度、弹性模量、拉伸强度和冲击强度分别为30.11MPa、2636MPa、14.65MPa和3.72kJ/m^(2),氧指数达27.3%,属难燃级别;与单独加APP时相比,力学性能降低,烟释放速总量降低11.93%,CO产率峰值和平均CO产率分别降低80.6%和49.3%,总失重率从63.53%降至61.42%,热稳定性能提高;FWPC燃烧后炭层表面凹凸不平,复合材料的阻燃和抑烟性能得到提升。

Optimizing Methylene Blue Removal from Textile Effluents: Comparative Study of Adsorption Efficiency Using Raw and Activated Carbon Derived from Gmelina Wood Wastes

This research investigates the efficacy of activated Gmelina Wood Sawdust (GWS) as an adsorbent for the removal of methylene blue (MB) dye from aqueous solutions, in comparison with raw GWS. The study employs laboratory experiments to assess the percentage of dye removal across various temperature and pH conditions. The adsorption process is scrutinized under different parameters, encompassing contact time, initial dye concentration, adsorbent dosage, temperature, and pH. Results demonstrate that activated GWS surpasses its raw counterpart, showcasing superior MB dye removal percentages. Extended contact times increased initial dye concentrations, and higher adsorbent dosages contribute positively to removal efficiency, while temperature exhibits an inverse relationship with dye removal. Optimal adsorption occurs at a pH of 7.0, aligning with the adsorbent’s zero-point charge (pHzpc), underscoring the role of surface charge in the adsorption process. This study underscores the potential of activated GWS as an economical and promising adsorbent material for addressing pollutants. Furthermore, the utilization of activated carbon derived from abundant agricultural waste underscores an environmentally conscious approach to adsorption applications. The ability to tailor the size and properties of activated carbon particles opens avenues for optimizing adsorption capabilities, thereby presenting opportunities for enhanced water treatment solutions.

Xanthan Gum—Bio-Based Raw Material for Wood Adhesive

Due to their lower environmental impact, ease of accessibility, low cost, and biodegradability, bio-renewable sources have been used extensively in the last several decades to synthesize adhesives, substituting petrochemical-based adhesive. Vegetable oils (including palm, castor, jatropha, and soybean oils), lactic acid, potato starch, and other bio-renewable sources are all excellent sources for the synthesis of adhesives that are being taken into consideration for the synthesis of “eco-friendly” adhesives. Due to their widespread use, accessibility, affordability, and biodegradability, biobased raw materials like carbohydrates used to synthesize wood and wood composite adhesive have gradually replaced petrochemical-based adhesive. Recently, xanthan gum, a naturally occurring polymer, has drawn the interest of scientists as a potentially petroleum source replacement. It possesses specific rheological characteristics, excellent water solubility, and stability to heat, and can be used as a binder, thickener, suspending agent, and stabilizer. Xanthan gum increases the adhesive strength in addition to increasing the viscosity of water-soluble adhesives. This article discusses xanthan gum as a potential substitute for traditional raw materials derived from petroleum that is used as a raw material for adhesives.

注意力可变形卷积网络的木质板材瑕疵识别

为了解决木材缺陷检测中人工成本高、效率低的问题,该文基于可变性卷积网络和注意力机制,提出一种端到端的神经架构模型.首先,可变形卷积网络(Deformable Convolutional Network, DCN)通过将矩形网格转换为变形网格,使模型专注于具有更多有用图像信息的区域.使用可变形卷积网络可以忽略图像特征中不相关的系数,解决了传统卷积在特征中学习更多信息能力有限的问题.然后,将DCN输出馈送到门控循环单元(Gated Recurrent Unit, GRU)层以学习缺陷图像的高级特征.最后,通过关注输入图像的最重要特征,应用注意力机制加强瑕疵区域的高亮度,从而提高模型识别的准确性.使用Matlab平台在4个木质板材缺陷数据集上将该文方法与现有其他方法进行比较分析,该文方法的准确率比其他3种对比方法提高了2.4%~13.2%的维度,灵敏度提高了3.3%~16.6%的维度,特异性提高了4%~21%的维度.实验结果表明,该文方法在检测精度和其他各个性能方面均优于现有方法,最佳准确率为99.2%,证明了该文方法的有效性.

Wood Density Determination with the Perspective to Decarbonisation of Tropical Forest Species from the Luki Biosphere Reserve in the Democratic Republic of the Congo

Specimens of the forest species such as Pentaclethra macrophylla, Petersianthus macrocarpus, Pycnanthus angolensis and Terminalia superba have been sampled from LUKI Biosphere reserve in the Democratic Republic of the Congo in order to determine their wood density with the perspective to decarbonisation. These parameters have been found out experimentally utilizing a drying technique in an oven including techniques of immersion in an Erlenmeyer full of water. The corresponding results indicated that the four species wood density is respectively 0.85, 0.80, 0.77 and 0.51. These preliminary results will be useful in our ongoing project on carbon dioxide absorption capacity of Congo rainforest tree species.

Environmentally Friendly Chitosan-Based Wood/Wood Composite Adhesive: Review

Due to the world’s energy issues and dependency on petroleum resources, focus has switched to finding new, sustainable raw material sources for wood adhesives. Renewable biopolymers would gradually replace petroleum and natural gas as the primary raw materials used in wood adhesives. Chitosan is a biomass substance having a lot of reserves. Chitosan is one of the most fascinating biopolymers in the adhesive sector because of its potential qualities for adhesive applications, such as biodegradability, biocompatibility, and non-toxicity. Chitosan and its derivatives have so garnered considerable interest in a wide range of adhesive applications. However, its adhesive strength is insufficient to glue wood under normal, humid conditions. There has been a lot of study done on how to make chitosan-based adhesives more cohesive and water resistant. In order to effectively use chitosan-based wood glue in wood/wood composite adhesive that gives comparable performance to synthetic adhesives, numerous new ways have been developed. It has been modified by the addition of various cross-linkers, including aldehydes like glyoxal glutaraldehyde etc., epoxy compounds, blended with other polymers, different acids and chitosan grafted onto vinyl acetate. In the production of wood composites, chitosan can also scavenge formaldehyde. This review of chitosan-based adhesives focuses on various cross-linkers for chitosan modification in order to improve the properties of chitosan-based wood adhesives.

Radio Frequency Gluing Technique for Wood-to-Wood Bonding: Review

White glues, also referred to as polyvinyl acetate (PVAc) adhesives, are thermoplastics that soften when heated to a specific degree and re-harden when cooled. For general assembly tasks in the joinery business, such as edge gluing, wood veneer, and edge bonding, white glues have been used. White glue requires a prolonged setting time to achieve adequate handling strength as it solidifies through evaporation or absorption of water by the wood. Cold press and hot press techniques are used to bond wood substrates. Recently, white glue is frequently used as a wood adhesive with dielectric heating systems to prevent this and enhance production speeds. Radio frequency (RF) curing is merely a technique for heating glue lines in wood-to-wood joints. It considered a source of heat, like steam and electricity. In order to bond wood substrates with the least amount of clamping time, the RF produces a very rapid, uniform rise in temperature. In this review paper, we discussed the radio frequency curing technique, their mechanism and troubleshooting to achieve perfect wood bond in joinery segment.

Comparative Study of Effect of Addition of Calcium Carbonate and Clay on the Performance Properties of Polyvinyl Acetate Wood Glue

Polyvinyl alcohol (PVA) stabilized Polyvinyl acetate (PVAc) dispersions-based wood adhesive has poor water and heat resistance. Recently, the addition of fillers in the wood adhesive is one of the most effective ways to enhance the performance of PVAc wood adhesive. Inorganic fillers have unique characteristics to improve the performance of adhesive, such as small size, high surface energy and surface hardness. Hence, the present work investigates the applicability of calcium carbonate and clay incorporated 3% in situ emulsion polymerization PVAc wood adhesive. Effect on physical, thermal and mechanical properties was studied by viscosity, pH, contact angle measurement, differential scanning calorimetry (DSC) and pencil hardness test of films. Emulsions with 3% calcium carbonate and 3% clay were prepared and the shear strength of the applied adhesive on wood was measured. The viscosity of the adhesives was reduced in the case of the addition of calcium carbonate and increased in the case of clay. The mechanical properties like tensile strength of adhesives with calcium carbonate and clay were measured by a universal tensile machine (UTM). Thermal stability was studied by differential scanning calorimetry (DSC). The tensile shear strength demonstrates that clay can improve bonding strength as compared to calcium carbonate of PVAc adhesive in wet conditions. The hardness of PVAc films was also changed positively by the addition of calcium carbonate and clay. Thermal stability of PVAc was significantly improved as calcium carbonate and clay were added to PVAc. Here, we did a comparative study of the effect of the addition of calcium carbonate and clay filler materials in situ polymerization of PVAc on their different properties.

高延性混凝土加固单层砖木结构农房振动台试验研究

为提高我国农村砌体房屋的抗震性能水平,采用高延性混凝土(HDC)条带加固的方法,设计了两个缩尺比为1∶2的砖木结构模型,对加固模型与未加固模型进行振动台同台对比试验。对比分析了地震作用下两个模型结构的加速度响应、位移响应、扭转效应和破坏形态,并对其抗震性能进行评估。试验结果表明:单层砖木结构农房的整体抗震性能较差,结构的自振频率低、位移反应较大,扭转效应明显,在8度设防地震作用下接近倒塌;经HDC条带加固的模型结构自振频率增大,位移反应相比未加固模型更小,山墙外闪现象得到明显控制,结构扭转效应的破坏作用明显减轻,在9度罕遇地震作用下HDC面层只出现轻微受损,可达到“大震可修”的性能目标。采用HDC条带加固法可有效延缓结构在强震作用下的刚度退化,增加结构的耐损伤能力,并有效提高此类砌体结构的整体抗震能力。

基于实验型和工业型近红外光谱仪的木片材性预测建模对比研究

本研究探讨了采用工业型光谱仪和实验型光谱仪建立近红外模型用于快速预测木片材性的可行性,分析对比不同光谱仪的建模效果。针对不同材性指标分析特点,通过光谱信号校正和特征波长筛选算法优化模型预测性能,建立适宜的建模方法。结果表明,基于工业型光谱仪建立的近红外模型对木片水分、基本密度、纤维形态等物理性能指标表现出更高的预测精度,而化学组分含量的预测精度与实验型光谱仪无显著差异。工业型光谱仪木片理化材性近红外模型适用于制浆原料品质评价和工业过程分析。

木材气干密度与基本密度关系模型比较

木材密度包括基本密度、气干密度等,在12%含水率条件下的气干密度(D12)较常用,因此有必要将木材气干密度换算为基本密度(Db)。目前利用木材气干密度计算基本密度的模型有Reyes、Chave、Simpson和Vieilledent模型等,然而这些模型预测结果不完全一致。利用中国林业科学研究院木材工业研究所(Research Institute of Wood Industry,Chinese Academy of Forestry,CRIWI)和法国农业国际合作研究发展中心(French Agricultural Research Centre for International Development,CIRAD)的木材D12和Db数据,首先基于CRIWI的木材密度数据建立D12与Db的关系模型,然后将CRIWI和CIRAD的D12数据分别代入Reyes模型、Chave模型、Simpson模型、Vieilledent模型和新建模型,获得每个树种木材Db的预测值,并根据Db预测值和实测值计算残差绝对值均值。不同模型残差绝对值均值比较结果表明:Reyes模型在利用CRIWI和CIRAD的木材密度数据时预测Db的准确性都比较高,适用性最广;Simpson模型、新建模型在D12高于1.0 g/cm3时预测Db的准确性降低。

南方型黑杨生长性状和木材材性与湿心材形成的相关性

为了分析南方型黑杨生长性状、木材性状对其湿心材形成的影响程度,于2023年11月份,在岳阳市君山区丁字堤南方型黑杨试验林,选取13年生的8个南方型黑杨无性系为研究对象,每个无性系选取6株样株,用9 mm生长锥在树高1.3 m处钻取木样,分别测量木芯长度、湿心材长度,并称量质量;依据试验设计,测定样木的胸径、树高、材积、树液pH、生材含水量、气干材含水量、木材基本密度;运用相关分析法、通径分析法,计算各性状对湿心材比例的直接影响、间接影响、决定系数,分析南方型黑杨无性系生长性状、木材性状的7个性状指标对湿心材比例的影响。结果表明:8个无性系间,湿心材比例、胸径、树高、材积、生材含水量存在极显著差异。树液pH和生材含水量,与湿心材比例存在极显著正相关,气干材含水量与湿心材比例存在显著正相关。通径分析表明,生材含水量(W r)、树液pH、气干材含水量(W a)是影响湿心材比例(R(wh))的重要因子,通过多元回归分析建立3个性状与湿心材比例的多元回归方程R(wh)=-0.995+0.006pH+0.075 W r+0.005 W a;其中,生材含水量对湿心材比例的直接影响(0.522)和单独决定系数大小表明,生材含水量对湿心材比例影响最大。因此,生材含水量大小可以作为选育南方型黑杨低湿心材选育时的评定指标。

不同热处理条件下橡胶木淀粉含量与抗蠹虫性能关系研究

为克服橡胶木易发生蠹虫危害的缺陷,采用不同热处理温度(90、120、150℃和180℃)和不同热处理时间(2、4 h和6 h)对橡胶木进行热处理,利用分光光度计考察不同温度和时间条件下热处理橡胶木中的淀粉含量变化,并分析蠹虫对橡胶木的蛀蚀程度与淀粉含量之间的关系。结果表明:热处理温度和处理时间对橡胶木中的淀粉含量有显著影响,且随着热处理温度的升高和热处理时间的延长,淀粉含量逐渐下降,从12.79%降至1.25%。在橡胶木抗蠹虫试验中,热改性处理后橡胶木的抗蠹虫效果均优于对照材,且橡胶木淀粉含量越低,其抗蠹虫效果越好。在实际生产过程中,在120℃下处理6 h为生产热处理橡胶木的优化条件,既能刚好维持橡胶木抗蠹虫的最低要求,也能降低热处理过程中木材强度的损失。