Re-usable Cd_(0.9)Zn_(0.1)S-ZnO@C/PVDF piezo-photocatalytic film with exceptional hydrogen evolution capability triggered by the synergetic advantages of piezoelectricity and S-Scheme heterojunction

Piezoelectric materials have advantages of fine-tuning photocatalytic performance through harvesting mechanical energy and open a new avenue in facilitating green catalytic reaction.Herein,polyvinylidene fluoride(PVDF),a flexible piezoelectric material,was introduced to synthesize a novel Cd_(0.9)Zn_(0.1)S-ZnO@C/PVDF(CZS-ZO@C/PVDF)piezo-photocatalytic film by spin coating and immersion phase conversion method.Benefiting from the piezoelectricity of PVDF and the internal electric field(IEF)of CZS-ZO@C Step-scheme(S-Scheme)heterojunction,CZS-ZO@C/PVDF was able to induce a hydrogen generation rate of 34.9 mmol g^(−1)h^(−1)activated by ultrasound and visible light(U-L),which is∼17.5 times of Cd_(0.9)Zn_(0.1)S/PVDF(CZS/PVDF)and∼7.4 times of the photocatalysis rate activated by visible light only(L).Piezoelectric measurements and COMSOL simulation illustrated the excellent piezoelectricity of CZS-ZO@C/PVDF film,which exhibits a piezoelectric coefficient(d33)of 9.9 pm V−1 and a piezoelectric potential of 874 mV(under 0.5 MPa).The reaction mechanism for the exceptional piezo-photocatalytic performance was finally disclosed through density functional theory(DFT)calculation and electrochemical tests.This study enriches the application scope of piezoelectric materials in sustainable energy catalysis and provides a new direction to develop efficient piezoelectric photocatalysts.

Enhancing Photocatalytic Hydrogen Evolution through Electronic Structure and Wettability Adjustment of ZnIn_(2)S_(4)/Bi_(2)O_(3) S-Scheme Heterojunction

The production of renewable fuels through water splitting via photocatalytic hydrogen production holds significant promise.Nonetheless,the sluggish kinetics of hydrogen evolution and the inadequate water adsorption on photocatalysts present notable challenges.In this study,we have devised a straightforward hydrothermal method to synthesize Bi_(2)O_(3)(BO)derived from metal‐organic frameworks(MOFs),loaded with flower-like ZnIn_(2)S_(4)(ZIS).This approach substantially enhances water adsorption and surface catalytic reactions,resulting in a remarkable enhancement of photocatalytic activity.By employing triethanolamine(TEOA)as a sacrificial agent,the hydrogen evolution rate achieved with 15%(mass fraction)ZIS loading on BO reached an impressive value of 1610μmol∙h^(−1)∙g^(−1),marking a 6.34-fold increase compared to that observed for bare BO.Furthermore,through density functional theory(DFT)and ab initio molecular dynamics(AIMD)calculations,we have identified the reactions occurring at the ZIS/BO S-scheme heterojunction interface,including the identification of active sites for water adsorption and catalytic reactions.This study provides valuable insights into the development of high-performance composite photocatalytic materials with tailored electronic properties and wettability.

Molten salt construction of core-shell structured S-scheme CuInS_(2)@CoS_(2) heterojunction to boost charge transfer for efficient photocatalytic CO_(2) reduction

Weak redox ability and severe charge recombination pose significant obstacles to the advancement of CO_(2) photoreduction.To tackle this challenge and enhance the CO_(2) photoconversion efficiency,fabricating well-matched S-scheme heterostructure and establishing a robust built-in electric field emerge as pivotal strategies.In pursuit of this goal,a core-shell structured CuInS_(2)@CoS_(2)S-scheme heterojunction was meticulously engineered through a two-step molten salt method.This approach over the CuInS_(2)-based composites produced an internal electric field owing to the disparity be-tween the Fermi levels of CoS_(2) and CuInS_(2) at their interface.Consequently,the electric field facili-tated the directed migration of charges and the proficient separation of photoinduced carriers.The resulting CuInS_(2)@CoS_(2) heterostructure exhibited remarkable CO_(2) photoreduction performance,which was 21.7 and 26.5 times that of pure CuInS_(2) and CoS_(2),respectively.The S-scheme heterojunc-tion photogenerated charge transfer mechanism was validated through a series of rigorous anal-yses,including in situ irradiation X-ray photoelectron spectroscopy,work function calculations,and differential charge density examinations.Furthermore,in situ infrared spectroscopy and density functional theory calculations corroborated the fact that the CuInS_(2)@CoS_(2) heterojunction substan-tially lowered the formation energy of *COOH and *CO.This study demonstrates the application potential of S-scheme heterojunctions fabricated via the molten salt method in the realm of ad-dressing carbon-related environmental issues.

Chalcogenide-based S-scheme heterojunction photocatalysts

The unique photocatalytic mechanism of S-scheme heterojunction can be used to study new and efficient photocatalysts.By carefully selecting semiconductors for S-scheme heterojunction photocatalysts,it is possible to reduce the rate of photogenerated carrier recombination and increase the conversion efficiency of light into energy.Chalcogenides are a group of compounds that include sulfides and selenides(e.g.,CdS,ZnS,Bi_(2)S_(3),MoS_(2),ZnSe,CdSe,and CuSe).Chalcogenides have attracted considerable attention as heterojunction photocatalysts owing to their narrow bandgap,wide light absorption range,and excellent photoreduction properties.This paper presents a thorough analysis of S-scheme heterojunction photocatalysts based on chalcogenides.Following an introduction to the fundamental characteristics and benefits of S-scheme heterojunction photocatalysts,various chalcogenide-based S-scheme heterojunction photocatalyst synthesis techniques are summarized.These photocatalysts are used in numerous significant photocatalytic reactions,in-cluding the reduction of carbon dioxide,synthesis of hydrogen peroxide,conversion of organic matter,generation of hydrogen from water,nitrogen fixation,degradation of organic pollutants,and sterilization.In addition,cutting-edge characterization techniques,including in situ characterization techniques,are discussed to validate the steady and transient states of photocatalysts with an S-scheme heterojunction.Finally,the design and challenges of chalcogenide-based S-scheme heterojunction photocatalysts are explored and recommended in light of state-of-the-art research.

Vacancy engineering mediated hollow structured ZnO/ZnS S-scheme heterojunction for highly efficient photocatalytic H_(2) production

Designing a step-scheme(S-scheme)heterojunction photocatalyst with vacancy engineering is a reliable approach to achieve highly efficient photocatalytic H_(2)production activity.Herein,a hollow ZnO/ZnS S-scheme heterojunction with O and Zn vacancies(VO,Zn-ZnO/ZnS)is rationally constructed via ion-exchange and calcination treatments.In such a photocatalytic system,the hollow structure combined with the introduction of dual vacancies endows the adequate light absorption.Moreover,the O and Zn vacancies serve as the trapping sites for photo-induced electrons and holes,respectively,which are beneficial for promoting the photo-induced carrier separation.Meanwhile,the S-scheme charge transfer mechanism can not only improve the separation and transfer efficiencies of photo-induced carrier but also retain the strong redox capacity.As expected,the optimized VO,Zn-ZnO/ZnS heterojunction exhibits a superior photocatalytic H_(2) production rate of 160.91 mmol g^(-1)h^(-1),approximately 643.6 times and 214.5 times with respect to that obtained on pure ZnO and ZnS,respectively.Simultaneously,the experimental results and density functional theory calculations disclose that the photo-induced carrier transfer pathway follows the S-scheme heterojunction mechanism and the introduction of O and Zn vacancies reduces the surface reaction barrier.This work provides an innovative strategy of vacancy engineering in S-scheme heterojunction for solar-to-fuel energy conversion.

Interfacial coordination bonds accelerate charge separation for unprecedented hydrogen evolution over S-scheme heterojunction

Inspired by natural photosynthesis,fabricating high-performance S-scheme heterojunction is regarded as a successful tactic to address energy and environmental issues.Herein,NH_(2)-MIL-125(Ti)/Zn_(0.5)Cd_(0.5)S/NiS(NMT/ZCS/NiS)S-scheme heterojunction with interfacial coordination bonds is successfully synthesized through in-situ solvothermal strategy.Notably,the optimal NMT/ZCS/NiS S-scheme heterojunction exhibits comparable photocatalytic H_(2)evolution(PHE)rate of about 14876.7μmol h^(−1)g^(−1)with apparent quantum yield of 24.2%at 420 nm,which is significantly higher than that of recently reported MOFs-based photocatalysts.The interfacial coordination bonds(Zn–N,Cd–N,and Ni–N bonds)accelerate the separation and transfer of photogenerated charges,and the NiS as cocatalyst can provide more catalytically active sites,which synergistically improve the photocatalytic performance.Moreover,theoretical calculation results display that the construction of NMT/ZCS/NiS S-scheme heterojunction also optimize the binding energy of active site-adsorbed hydrogen atoms to enable fast adsorption and desorption.Photoassisted Kelvin probe force microscopy,in-situ irradiation X-ray photoelectron spectroscopy,femtosecond transient absorption spectroscopy,and theoretical calculations provide sufficient evidence of the S-scheme charge migration mechanism.This work offers unique viewpoints for simultaneously accelerating the charge dynamics and optimizing the binding strength between the active sites and hydrogen adsorbates over S-scheme heterojunction.

S型异质结光催化剂ZnFe_(2)O_(4)/WO_(3)的构筑及光催化还原CO_(2)性能

通过在WO_(3)纳米片表面负载ZnFe_(2)O_(4)纳米颗粒,构建了一系列S型异质结光催化剂ZnFe_(2)O_(4)/WO_(3),并研究了其光催化CO_(2)还原性能。在没有助催化剂和牺牲剂的条件下,所制备的ZnFe_(2)O_(4)/WO_(3)复合材料可对CO_(2)与水蒸汽进行光催化反应。优化后的材料光照5 h后CO_(2)还原产物CO和CH_(4)的产量分别为7.87和4.88μmol·g^(-1)。相对于单相组分,CO和CH_(4)的产量明显提高。光催化活性的提高,归因于ZnFe_(2)O_(4)和WO_(3)异质结的形成以及光生载流子的S型电荷传输模式。

宽负荷下切圆燃煤锅炉H_(2)S分布特性的数值模拟

锅炉采用空气分级燃烧降低NO_(x)排放的同时也提高了主燃区H_(2)S体积分数。炉墙壁面过高的H_(2)S体积分数是加剧水冷壁高温腐蚀的重要因素。为保障新能源并网发电,大型燃煤机组灵活调峰的需求增加,不同负荷下的水冷壁近壁面H_(2)S分布特性值得关注。通过正交试验分析了切圆燃煤锅炉运行参数对水冷壁近壁面H_(2)S体积分数分布的影响。选取一台超临界600 MW切圆燃煤锅炉建立数值模型,设计L_(16)(4^(5))正交工况,覆盖100%BMCR、75%THA,50%THA以及35%BMCR四种负荷。建立了自定义SO_(x)生成模型以确定燃料硫的析出和转化路径,模型包含多表面反应子模型以描述焦炭与O_(2)/CO_(2)/H_(2)O等3种气体的异相反应,并确定焦炭气化反应消耗量占总消耗量的比例,进而对炉膛H_(2)S空间分布进行了模拟计算。研究表明,近壁面高体积分数H_(2)S区域主要位于投运燃烧器层中最下层燃烧器以下以及最上层燃烧器以上至SOFA层之间,烟气切圆沿炉膛高度增加逐渐增大是造成后一区域H_(2)S体积分数较高的重要原因。35%BMCR负荷下水冷壁重点区域的H_(2)S平均体积分数为364μL/L,明显低于其他负荷。锅炉运行参数对重点区域H_(2)S体积分数影响程度的排序为:锅炉负荷>一次风率>主燃区空气过量系数>假想切圆直径>燃烧器竖直摆角。

结合主动光源和改进YOLOv5s模型的夜间柑橘检测方法

【目的】解决夜间环境下遮挡和较小柑橘难以准确识别的问题,实现采摘机器人全天候智能化作业。【方法】提出一种结合主动光源的夜间柑橘识别方法。首先,通过分析主动光源下颜色特征不同的夜间柑橘图像,选择最佳的光源色并进行图像采集。然后,提出一种夜间柑橘检测模型BI-YOLOv5s,该模型采用双向特征金字塔网络(Bi-FPN)进行多尺度交叉连接和加权特征融合,提高对遮挡和较小果实的识别能力;引入Coordinate attention(CA)注意力机制模块,进一步加强对目标位置信息的提取;采用融入Transformer结构的C3TR模块,在减少计算量的同时更好地提取全局信息。【结果】本文提出的BI-YOLOv5s模型在测试集上的精准率、召回率、平均准确率分别为93.4%、92.2%和97.1%,相比YOLOv5s模型分别提升了3.2、1.5和2.3个百分点。在所采用的光源色环境下,模型对夜间柑橘识别的正确率为95.3%,相比白光环境下提高了10.4个百分点。【结论】本文提出的方法对夜间环境下遮挡和小目标柑橘的识别具有较高的准确性,可为夜间果蔬智能化采摘的视觉精准识别提供技术支持。

基于CED-YOLOv5s模型的煤矸识别方法研究

由于煤矿井下高噪声、低照度、运动模糊的复杂工况和煤矸易聚集现象,导致煤矸目标检测模型特征提取困难及煤矸分类、定位不准确问题。针对该问题,提出一种基于CED-YOLOv5s模型的煤矸识别方法。首先,在YOLOv5s主干网络中引入坐标注意力(CA)机制,通过将坐标信息嵌入信道关系和长程依赖关系中对特征图进行编码,充分利用通道注意力信息和空间注意力信息,使模型更加关注重要特征,抑制无用信息。其次,在YOLOv5s的检测头部引入EIoU回归损失函数,将目标框与锚框的宽高差异最小化,以增强目标的位置和边界信息,提高模型在密集目标下的定位精度和收敛速度;最后,在YOLOv5s的检测头部引入轻量化解耦头,解耦出单独的特征通道,分别用于分类任务和回归任务,解决了原模型中耦合头部分类任务与回归任务的相互干扰问题,进一步提升了模型的并行运算效率与检测精度。实验结果表明:CED-YOLOv5s模型与其他YOLO系列目标检测模型相比,综合性能最佳,平均检测精度达94.8%,相较于YOLOv5s模型提升了3.1%,检测速度达84.8帧/s,可充分满足煤矿井下煤矸实时检测需求。

基于改进YOLOv5s的不同成熟度苹果目标检测方法

[目的]本文旨在解决在自然环境下不同成熟度苹果目标检测精度较低的问题。[方法]提出了一种改进的YOLOv5s模型SODSTR-YOLOv5s(YOLOv5s with small detection layer and omni-dimensional dynamic convolution and swin transformer block),用于不同成熟度苹果检测。首先改进YOLOv5s的多尺度目标检测层,在Prediction中构建检测160×160特征图的检测头,提高小尺寸的不同成熟度苹果的检测精度;其次在Backbone结构中融合Swin Transformer Block,加强同级成熟度的苹果纹理特征融合,弱化纹理特征分布差异带来的消极影响,提高模型泛化能力;最后将Neck结构的Conv模块替换为动态卷积模块ODConv,细化局部特征映射,实现局部苹果细粒度特征的充分提取。基于不同成熟度苹果数据集进行试验,验证改进模型的性能。[结果]改进模型SODSTR-YOLOv5s检测的精确率、召回率、平均精度均值分别为89.1%、95.5%、93.6%,高、中、低成熟度苹果平均精度均值分别为94.1%、93.1%、93.7%,平均检测时间为16 ms,参数量为7.34 M。相比于YOLOv5s模型,改进模型SODSTR-YOLOv5s精确率、召回率、平均精度均值分别提高了3.8%、5.0%、2.9%,参数量和平均检测时间分别增加了0.32 M和5 ms。[结论]改进模型SODSTR-YOLOv5s提升了在自然环境下对不同成熟度苹果的检测能力,能较好地满足实际采摘苹果的检测要求。

基于改进YOLOv5s的小目标检测算法

针对当前主流目标检测算法对图像中远距离小目标产生的漏检、误检等问题,提出一种改进YOLOv5s的小目标检测算法。在模型训练过程中,通过引入Focal-EIOU定位损失函数,加强边界框的定位精度;在骨干网络中,通过添加小目标检测层,提高小目标的检测精度;在Neck结构中,通过优化上采样算子和添加注意力机制,加强小目标的特征信息。实验结果表明,改进后的算法在VisDrone数据集上与YOLOv5s算法相比,mAP@small提高了3.2%,且检测速度满足实时性的要求,能够很好地应用于小目标检测任务中。

σ相对S32205双相不锈钢力学性能和耐蚀性的影响

通过JMatPro软件模拟计算和试验结合的方式研究了S32205双相不锈钢中σ相的析出行为及对其力学性能和耐蚀性的影响。结果表明:通过JMatPro软件模拟计算得到的S32205双相不锈钢σ相析出鼻尖温度为875℃。当S32205双相不锈钢在875℃保温0~90 min,σ相含量快速增加,当保温时间增加至90~120 min时,σ相含量缓慢上升,而当保温时间为120~240 min时σ相含量达到饱和,最大为6.1%,与模拟计算结果基本符合。随着保温时间逐渐增加,S32205双相不锈钢的屈服强度、抗拉强度及显微硬度先快速增加后逐渐保持稳定,而伸长率呈下降趋势,强度和硬度与σ相体积含量成正比关系,σ相强化机理由Orowan强化机制转变为σ相的高弹性应变能强化机制。不同σ含量的S32205双相不锈钢的电化学试验结果表明随着σ相的含量逐渐增多,基体越容易贫Cr, S32205双相不锈钢越容易产生腐蚀,耐蚀性逐渐降低。

基于CFS-YOLO算法的复杂工况环境下煤矸图像识别方法

针对煤矿高噪声、低照度、运动模糊与大批量煤矸混杂等复杂工况环境因素导致煤矸识别存在误检、漏检以及检测精度低的问题,提出一种基于CFS-YOLO算法的煤矸智能识别模型。采用ConvNeXt V2(Convolutional Neural Network with NeXt Units Version 2)特征提取模块替换主干网络末端的2个C3(Cross Stage Partial Bottle Neck Mudule)模块,通过将掩码自动编码器(Masked Autoencoders,MAE)和全局响应归一化(Global Response Normalization,GRN)层添加到ConvNeXt架构中,有效缓解特征崩溃问题以及保持特征在网络传递过程中的多样性;采用Focal-EIOU(Focal and Efficient Intersection Over Union)损失函数替换原CIOU(Computer Intersection Over Union)损失函数,通过其Focal-Loss机制和调整样本权重的方式优化边界框回归任务中的样本不平衡问题,提高模型的收敛速度和定位精度;添加无参注意力机制(Simple Attention Mechanism,SimAM)于主干网络每个C3模块的后端,凭借其注意力权重自适应调整策略,提升模型对尺度变化较大或低分辨率煤矸目标关键特征的提取能力。通过消融试验和对比试验验证所提CFS-YOLO模型的有效性与优越性。试验结果表明:CFS-YOLO模型对于煤矸在煤矿高噪声、低照度、运动模糊与大批量煤矸混杂等复杂环境下的检测效果均得到有效提高,模型的平均精度均值达到90.2%,相较于原YOLOv5s模型的平均精度均值提高了3.7%,平均检测速度达到90.09 FPS,可充分满足煤矸实时检测的需求。同时与YOLOv5s、YOLOv7-tiny与YOLOv8n等6种YOLO系列算法相比,CFS-YOLO模型对煤矿复杂环境的适应性最强且综合检测性能最佳,可为煤矸的智能高效分选提供技术支持。

S31603复合板压力容器点蚀行为试验研究

针对S31603复合板压力容器不锈钢层不同点蚀深度的发展进程及穿透至碳钢层后的电偶腐蚀风险,通过预制不同点蚀深度的S31603复合板试片组,分别在实验室和现场S31603复合板压力容器内进行腐蚀试验,并对试片进行宏观分析、低倍观察、高倍观察及腐蚀产物表征分析。结果表明,S31603复合板预制的点蚀坑深度穿透至碳钢层时发生了明显的电偶腐蚀,实验室腐蚀试验和现场试验最高点蚀速率分别达到4.954,1.023 mm/a;用S31603复合板试片预制的点蚀坑深度穿透至碳钢层后碳钢优先发生电偶腐蚀;S31603复合板试片预制的点蚀坑深度处于不锈钢层时腐蚀轻微,也未萌生新的点蚀,相邻点蚀坑之间也未见明显的腐蚀影响。在含Cl-的酸性腐蚀性介质环境下,当S31603复合板压力容器点蚀深度处于不锈钢层时腐蚀轻微,当点蚀深度穿透至碳钢层后,会发生明显的电偶腐蚀,需要缩短S31603复合板压力容器开罐检修周期以保证安全使用。

基于改进YOLOv5s的轻量级绝缘子缺失检测

针对现有绝缘子缺失检测模型计算复杂度高和小目标难以检测等问题,提出一种基于改进的YOLOv5s轻量级检测模型。首先,移除主干网络中的C3模块来减少模型的参数量。其次,在多尺度特征融合网络中引入卷积块注意力机制来提高复杂背景下模型的特征提取能力。同时,采用加权双向特征金字塔网络结构对特征进行双向跨尺度加权融合,提升网络在遮挡物、相似目标干扰下目标的检测性能。最后,选用SIoU损失函数提升网络的收敛速度和检测精度。实验结果表明,所提模型的平均精准率为96.8%,浮点运算数为2.8 GFLOPS,而原始YOLOv5s在保证97.4%的平均精准率下的浮点运算数为16.3 GFLOPS。相较于原始模型,所提模型对小目标、遮挡目标以及模糊等场景有着较强的鲁棒性,且在保证近似检测精度的同时极大减少了计算量。

基于YOLOv5s的改进实时红外小目标检测

针对红外图像分辨率低、背景复杂、目标细节特征缺失等问题,提出了一种基于YOLOv5s的改进实时红外小目标检测模型Infrared-YOLOv5s。在特征提取阶段,采用SPD-Conv进行下采样,将特征图切分为特征子图并按通道拼接,避免了多尺度特征提取过程中下采样导致的特征丢失情况,设计了一种基于空洞卷积的改进空间金字塔池化模块,通过对具有不同感受野的特征进行融合来提高特征提取能力;在特征融合阶段,引入由深到浅的注意力模块,将深层特征语义特征嵌入到浅层空间特征中,增强浅层特征的表达能力;在预测阶段,裁减了网络中针对大目标检测的特征提取层、融合层及预测层,降低模型大小的同时提高了实时性。首先通过消融实验验证了提出各模块的有效性,实验结果表明,改进模型在SIRST数据集上平均精度均值达到了95.4%,较原始YOLOv5s提高了2.3%,且模型大小降低了72.9%,仅为4.5 M,在Nvidia Xavier上推理速度达到28 f/s,利于实际的部署和应用。在Infrared-PV数据集上的迁移实验进一步验证了改进算法的有效性。提出的改进模型在提高红外图像小目标检测性能的同时,能够满足实时性要求,因而适用于红外图像小目标实时检测任务。

基于RFID和磁导航的AGV小车S7-1200PLC控制系统设计

该项目利用S7-1200PLC作为AGV主控制器,采取驱动轮差速进行转向控制,采用磁导航传感器和RFID标签传感器实现循迹和定位功能,实现AGV小车在立体仓库、VMC850数控加工中心、CK40S数控车床、RB20工业机器人、装配单元之间产品的搬运功能。

高校仪器设备10S精细化管理的研究与实践

高校仪器设备的投入不断增加,如何管好、管活仪器设备,使其在教学和科研方面发挥更大的作用,是广大高校仪器设备管理工作者努力解决的问题。四川大学经过长期积极探索与不断实践,首次提出10S精细化仪器设备管理体系,以管理促发展,理论与实践相结合,仪器设备日常管理规范化与现代化相结合,大幅提升了仪器设备管理的科学化和精细化水平,有效提高了仪器设备开放共享的效益和受益面。对10S精细化管理在仪器设备管理中的具体内容进行了阐述与分析,可为其他高校仪器设备管理提供一定的借鉴和帮助。

改进YOLOv5s对病理学图像中猪只小肠绒毛的检测

为解决传统小肠绒毛需要专业人员手动检测耗时耗力且存在主观性和不稳定性等问题,同时提高在复杂病理学图像中小肠绒毛检测的准确率和效率,该研究提出基于改进YOLOv5s检测复杂病理学图像下猪只小肠绒毛的方法。首先,采用串联形式的混合池化对空间金字塔进行优化,增强特征提取与特征表达,提升检测精度;然后引入一种基于注意力机制的网络模块(simpleattentionmechanism,SimAM)与Bottleneck中的残差连接相结合,使用SimAM对Bottleneck中的特征图进行加权,得到加权后的特征表示,利用注意力机制加强模型对目标的感知。试验结果表明,该研究算法的平均精度(average precision)和每秒传输帧数(frame per second,FPS)达到92.43%和40帧/s。改进后的YOLOv5s在召回率和平均精度上相较改进前提高2.49和4.62个百分点,在不增加模型参数量的情况下,每帧图片的推理时间缩短1.04 ms。与经典的目标检测算法SSD、Faster R-CNN、YOLOv6s、YOLOX相比,平均精度分别提高15.16、10.56、2.03和4.07个百分点。结果表明,该方法能够实现病理学图像中猪只小肠绒毛自动化检测,保证复杂图像检测速度的同时,提高了小肠绒毛的检测精度。