针对移动除草设备的YOLOv5轻量级网络优化与杂草检测方法

针对当前杂草检测算法普遍面临的模型参数量庞大、实时检测性能不佳以及难以适配移动除草设备等挑战,提出了一种高效且轻量化的网络模型。该模型的核心在于对YOLOv5框架的整体改造与优化,旨在实现资源占用最小化与检测性能的最优化平衡。具体而言,引入了基于Bottleneck设计的ShuffleNet V2网络作为YOLOv5的骨干架构,这一举措有效利用了ShuffleNet V2在保持高效推理速度的同时减少参数量的优势。进一步地,将原卷积模块与C3模块分别替换为CSconv模块和C3Ghost模块,这些轻量化组件不仅减轻了计算负担,还通过更高效的特征提取机制提升了模型性能。为增强模型在复杂多变环境中的目标识别能力,融入了无参数SimAM注意力机制,该机制无需额外学习参数即可动态调整特征图的重要性,从而增强了模型对关键信息的敏感度。同时,还集成了轻量级的空间组增强模块(SGE),该模块通过优化空间特征分布,进一步提升了模型的鲁棒性和准确性。为了验证改进后的模型能够适配移动设备,将此模型移植到Jetson nano开发板上进行试验,数据集为芝麻作物与杂草检测数据集。实验数据表明:经过这一系列改进后,新模型的参数量仅为原始YOLOv5网络的1/5,模型体积压缩至2.9MB,在开发板上的帧率为23fps,一般摄像头拍摄的视频帧率为25fps,基本可以在开发板上流畅运行,尽管模型准确率略有下降(仅0.09%),但这一微小牺牲换来了模型在资源受限条件下的卓越表现,实现了检测精度与计算效率的完美平衡。

JAK2V617F在原发性血小板增多症的表达情况与临床特征分析

目的通过对原发性血小板增多症(ET)患者进行JAK2 V617 F突变的检测,研究JAK2 V617 F突变对于ET患者的临床特征以及预后的关系,为ET患者诊断治疗提供帮助。方法按照世界卫生组织(WHO)诊断标准,收集2018年4月至2024年2月确诊的82例ET患者,通过收集其外周血血小板数、中性粒细胞数、白细胞数、血红蛋白以及纤维蛋白原含量,利用聚合酶链式反应(PCR)技术检测骨髓或外周血中JAK2 V617 F突变的表达。结果(1)本研究收集的82例ET患者中,存在JAK2 V617 F突变的患者有65例,占所有ET患者的79.27%;(2)在ET患者中,JAK2 V617 F基因的表达与性别无关(P=0.947),但是与年龄相关,>50岁组JAK2 V617 F基因突变占比更高,差异具有统计学意义(P=0.020);(3)存在JAK2 V617 F基因的突变的患者血小板数低于阴性患者,差异具有统计学意义(P=0.017),而中性粒细胞数、白细胞数以及纤维蛋白原的检出量则明显较高,差异均有统计学意义(P<0.001);(4)观察患者预后情况,JAK2 V617 F基因的突变会提升血栓以及脾增大风险,差异均有统计学意义(P<0.05)。结论JAK2V617 F突变对原发性血小板增多症的预后产生较大影响,同时可能参与原发性血小板增多症的发病过程。

罗汉果苷V调控高糖状态巨噬细胞M1极化促进骨髓间充质干细胞的成骨分化

背景:糖尿病微环境会造成巨噬细胞过度M1极化,这种高糖炎症状态会抑制骨髓间充质干细胞的成骨分化,从而影响糖尿病骨缺损的愈合。研究表明罗汉果苷Ⅴ具有抗炎、抗氧化、降血糖的作用,但其能否调节高糖炎症状态下巨噬细胞M1极化及骨髓间充质干细胞的成骨分化尚不清楚。目的:探讨罗汉果苷Ⅴ在高糖炎症状态下调节巨噬细胞M1型极化对骨髓间充质干细胞成骨分化的影响。方法:构建糖尿病C57BL/6小鼠模型,从正常和糖尿病小鼠分离骨髓来源巨噬细胞,分别培养于低糖和高糖培养基。使用脂多糖和干扰素γ作为炎症刺激诱导骨髓来源巨噬细胞的M1型极化,同时以160,320,640μmol/L罗汉果苷Ⅴ干预,用流式细胞术检测F4/80^(+)CD86^(+)细胞比例,qRT-PCR检测诱导型一氧化氮合酶、白细胞介素1β、白细胞介素6的mRNA表达水平,ELISA检测骨髓来源巨噬细胞上清液中肿瘤坏死因子α水平。分离C57BL/6小鼠骨髓间充质干细胞,分别使用低糖或高糖成骨诱导液诱导成骨分化,添加M1型巨噬细胞条件培养基作为炎症刺激,以及320μmol/L罗汉果苷Ⅴ干预,成骨诱导14 d后采用qRT-PCR检测碱性磷酸酶、Runt相关因子2、骨钙素、骨桥蛋白的mRNA表达水平,成骨诱导21 d后进行茜素红染色及定量分析。结果与结论:①流式细胞术结果显示320,640μmol/L罗汉果苷Ⅴ组的F4/80^(+)CD86^(+)细胞比例明显低于高糖炎症对照组(P<0.05);②qRT-PCR结果显示160,320,640μmol/L罗汉果苷Ⅴ组的诱导型一氧化氮合酶、白细胞介素6的mRNA相对表达量较高糖炎症对照组显著降低(P<0.05),320,640μmol/L罗汉果苷Ⅴ组白细胞介素1β的mRNA相对表达量较高糖炎症对照组显著降低(P<0.05);③ELISA结果显示160,320,640μmol/L罗汉果苷Ⅴ组的肿瘤坏死因子α分泌水平较高糖炎症对照组显著降低(P<0.05);④320μmol/L罗汉果苷Ⅴ干预后,高糖炎症状态下骨髓间充质干细胞的钙盐沉积增加(P<0.05),且碱性磷酸酶、Runt相关因子2和骨桥蛋白的mRNA相对表达量增加(P<0.05)。结果表明,罗汉果苷Ⅴ可通过抑制高糖炎症状态下骨髓来源巨噬细胞的M1型极化及炎症因子表达,促进骨髓间充质干细胞的成骨分化。

GRBAS和CAPE-V量表中文听感知评估一致性分析

目的 评估在中文语境下GRBAS和CAPE-V的评估者内和评估者间一致性。方法 5名嗓音相关专业人员使用《发音评估助手》APP对抽取自解放军总医院第六医学中心病理嗓音数据库V1.0的52例语音样本进行GRBAS和CAPE-V评估,52例用于评估者间一致性分析,其中的38例用于评估者内部一致性评估。使用ICC对CAPE-V进行评估者内和评估者间一致性分析,采用Cohen's Kappa方法对GRBAS的5个特征进行评估者内一致性分析,采用Fleiss Kappa方法对GRBAS评估者间一致性进行分析,并对两个量表中相同的特征进行Spearman相关性分析。结果 评估者内一致性分析结果示,CAPE-V 6个特征均呈现出高的一致性,各特征的ICC系数分别为总体音质(OS)=0.80,粗糙度(R)=0.69,气息音(B)=0.77,紧张度(S)=0.75,音调异常程度(P)=0.74,响度异常程度(L)=0.78;GRBAS除特征G(0.48)和S(0.45)存在较弱的评估者内一致性外,其余特征的评估者内一致性均较差。评估者间一致性分析结果示,CAPE-V各特征的ICC相关系数均大于0.85,提示不同评估者对各特征的评分均具有很高的一致性;GRBAS除G特征(相关系数=0.48)外,其余特征的相关系数均小于0.40,说明除特征G外,其余特征的评估者间一致性均较差。在同时使用GRBAS和CAPE-V对同一样本进行评估时,总体音质(OS/G)、粗糙度(R)、气息音(B)、紧张度(S)的Spearman相关系数分别为0.89、0.85、0.91、0.91,提示两个量表中的这4个特征的评分结果具有高度相关性。结论 在中文语音样本的听感知评估中,CAPE-V量表的评估者内和评估者间一致性更高,较GRBAS量表更适用于临床。

城区复杂条件下V形墩钢-混组合连续梁桥临时工程专项设计研究

本研究以双桥路道路工程施工为平台,开展城区复杂条件下V形墩钢-混组合连续梁桥支架体系研究。结合项目地形、地质特点及受限空间等多方面工况,利用三维建模及结构计算等手段综合考虑设计一套安全风险可控、经济价值高的临时工程体系,同时,结合施工工序,将下部结构V形墩、钢箱梁顶推、上部结构混凝土箱梁等主要结构的支架统筹考虑,周转利用,节约施工工期,降低施工成本。

Cohort study on the treatment of BRAF V600E mutant metastatic colorectal cancer with integrated Chinese and western medicine

BACKGROUND Patients with BRAF V600E mutant metastatic colorectal cancer(mCRC)have a low incidence rate,poor biological activity,suboptimal response to conventional treatments,and a poor prognosis.In the previous cohort study on mCRC conducted by our team,it was observed that integrated Chinese and Western medicine treatment could significantly prolong the overall survival(OS)of patients with colorectal cancer.Therefore,we further explored the survival benefits in the population with BRAF V600E mutant mCRC.AIM To evaluate the efficacy of integrated Chinese and Western medicine in the treatment of BRAF V600E mutant metastatic colorectal cancer.METHODS A cohort study was conducted on patients with BRAF V600E mutant metastatic colorectal cancer admitted to Xiyuan Hospital of China Academy of Chinese Medical Sciences and Traditional Chinese Medicine Hospital of Xinjiang Uygur Autonomous Region from January 2016 to December 2022.The patients were divided into two cohorts.RESULTS A total of 34 cases were included,with 23 in Chinese-Western medicine cohort(cohort A)and 11 in Western medicine cohort(cohort B).The median overall survival was 19.9 months in cohort A and 14.2 months in cohort B,with a statistically significant difference(P=0.038,hazard ratio=0.46).The 1-3-year survival rates were 95.65%(22/23),39.13%(9/23),and 26.09%(6/23)in cohort A,and 63.64%(7/11),18.18%(2/11),and 9.09%(1/11)in cohort B,respectively.Subgroup analysis showed statistically significant differences in median OS between the two cohorts in the right colon,liver metastasis,chemotherapy,and first-line treatment subgroups(P<0.05).CONCLUSION Integrated Chinese and Western medicine can prolong the survival and reduce the risk of death in patients with BRAF V600E mutant metastatic colorectal cancer,with more pronounced benefits observed in patients with right colon involvement,liver metastasis,combined chemotherapy,and first-line treatment.

构式“一V一个不吱声”的功用及流行动因

“一V一个不吱声”作为一种构式,自2023年便常见于网络及日常生活中。该结构具有凝固性强、对语境依赖性强且受事位置灵活的特点,其语义由最初对某一客观事实进行描述,逐渐泛化为对某种结果或某种程度进行主观评价。该构式产生及流行的动因在于强大的方言基础、模因的复制与传播、网络的便捷性、大众跟风效应的影响及礼貌原则的约束。

基于指令扩展的RISC-V可配置故障注入检测方法

针对处理器运行时受到故障攻击出现的数据流错误,提出了一种面向RISC-V处理器微架构的模式可配置故障注入检测方法。该方法基于RISC-V指令集架构,利用其可扩展性添加带模式配置的自定义算术逻辑运算指令和控制与状态寄存器,以软硬件结合的方式同时实现算术逻辑运算和故障注入检测。在软件层面,通过写寄存器指令将配置信息写入自定义的控制与状态寄存器,配置自定义指令的故障注入检测模式,包括信息冗余和时间冗余2种故障注入检测模式及其参数;在硬件层面,实现了支持模式可配置故障注入检测方法的RISC-V处理器微架构。采用仿真器命令模拟故障注入,验证扩展后的RISC-V处理器的功能正确性与故障注入检测能力。实验结果表明:当信息冗余模式和时间冗余模式的应用频率相同时,模式可配置方法相较于单信息冗余方法,平均故障检测率提高13.34%,引入4.4%的平均资源开销;相较于单时间冗余方法,降低了8.24%的平均时间开销,故障检测率降低了13.33%。所提模式可配置方法可以实现故障检测率和时间开销的折中,适用于不同安全性和性能需求的应用场景。

外泌体miR-47V2-3p在结直肠癌化疗耐药中的作用及机制研究

目的探讨外泌体miR-47V2-3p在结直肠癌化疗耐药中的作用及机制。方法纳入60只SD级小鼠,先建立结直肠癌模型,后将其分成氟尿嘧啶给药组、外泌体携带miR-47V2-3p模拟物组、外泌体携带miR-47V2-3p抑制剂组、氟尿嘧啶+外泌体携带miR-47V2-3p模拟物组、氟尿嘧啶+外泌体携带miR-47V2-3p抑制剂组和对照组6组各10只小鼠。小鼠均接种SW480细胞,通过原位杂交检测小鼠细胞miR-47V2-3p表达水平,同时检测SW480细胞增殖抑制率。结果与对照组比较,氟尿嘧啶给药组、外泌体携带miR-47V2-3p模拟物组、外泌体携带miR-47V2-3p抑制剂组、氟尿嘧啶+外泌体携带miR-47V2-3p模拟物组、氟尿嘧啶+外泌体携带miR-47V2-3p抑制剂组检测组织miR-47V2-3p表达水平和不同培养基浓度下SW480细胞增殖抑制率均升高,差异有统计学意义(P<0.05)。与氟尿嘧啶给药组比较,外泌体携带miR-47V2-3p模拟物组、外泌体携带miR-47V2-3p抑制剂组、氟尿嘧啶+外泌体携带miR-47V2-3p模拟物组、氟尿嘧啶+外泌体携带miR-47V2-3p抑制剂组检测组织miR-47V2-3p表达水平升高;与外泌体携带miR-47V2-3p模拟物组比较,氟尿嘧啶+外泌体携带miR-47V2-3p模拟物组、氟尿嘧啶+外泌体携带miR-47V2-3p抑制剂组检测组织miR-47V2-3p表达水平升高;与外泌体携带miR-47V2-3p抑制剂组比较,氟尿嘧啶+外泌体携带miR-47V2-3p模拟物组、氟尿嘧啶+外泌体携带miR-47V2-3p抑制剂组检测组织miR-47V2-3p表达水平升高;与氟尿嘧啶+外泌体携带miR-47V2-3p模拟物组比较,氟尿嘧啶+外泌体携带miR-47V2-3p抑制剂组检测组织miR-47V2-3p表达水平升高,差异均有统计学意义(P<0.05)。氟尿嘧啶给药组、外泌体携带miR-47V2-3p模拟物组、外泌体携带miR-47V2-3p抑制剂组、氟尿嘧啶+外泌体携带miR-47V2-3p模拟物组、氟尿嘧啶+外泌体携带miR-47V2-3p抑制剂组不同培养基浓度下SW480细胞增殖抑制率依次逐渐降低,两两比较差异均有统计学意义(P<0.05)。结论针对结直肠癌,外泌体miR-47V2-3p在化疗耐药中起着关键作用,可影响化疗的敏感性及效果。

Soil Moisture Retrieval from Satellite Images and Its Application to Heavy Rainfall Simulation in Eastern China

The soil water index （SWI） from satellite remote sensing and the observational soil moisture from agricultural meteorological stations in eastern China are used to retrieve soil moisture. The analysis of correlation coefficient （CORR）, root-mean-square-error （RMSE） and bias （BIAS） shows that the retrieved soil moisture is convincible and close to the observation. The method can overcome the difficulties in soil moisture observation on a large scale and the retrieved soil moisture may reflect the distribution of the real soil moisture objectively. The retrieved soil moisture is used as an initial scheme to replace initial conditions of soil moisture （NCEP） in the model MM5V3 to simulate the heavy rainfall in 1998. Three heavy rainfall processes during 13-14 June, 18-22 June, and 21-26 July 1998 in the Yangtze River valley are analyzed. The first two processes show that the intensity and location of simulated precipitation from SWI are better than those from NCEP and closer to the observed values. The simulated heavy rainfall for 21-26 July shows that the update of soil moisture initial conditions can improve the model＇s performance. The relationship between soil moisture and rainfall may explain that the stronger rainfall intensity for SWI in the Yangtze River valley is the result of the greater simulated soil moisture from SWI prior to the heavy rainfall date than that from NCEP, and leads to the decline of temperature in the corresponding area in the heavy rainfall days. Detailed analysis of the heavy rainfall on 13-14 June shows that both land-atmosphere interactions and atmospheric circulation were responsible for the heavy ralnfall, and it shows how the SWI simulation improves the simulation. The development of mesoscale systems plays an important role in the simulation regarding the change of initial soil moisture for SWI.

Effects of electron radiation on commercial power MOSFET with buck converter application

Microelectronic power converters such as buck and boost converter are required to be tolerant to radiations including electron radiation. This paper examines electron radiation effects on the Ⅰ-Ⅴ characteristics of VDMOSFET and its corresponding effects in buck converter. Analysis of the electrical characteristics shows that after irradiation the threshold voltage and drain current for all VDMOSFETs degraded more than two orders of magnitude. The impact of this electrical degradation has been investigated in an application of typical buck converter circuit. The buck converter with n-channel switching transistor shows that after irradiation its output voltage increased with the drain current in the n-channel ZVN4424 A VDMOSFET, while the buck converter with p-channel switching transistor shows its output voltage decreased with the drain current in the p-channel ZVP4424 A VDMOSFET after irradiation.

A New Approach for Manufacturing a High Porosity Ti-6Al-4V Scaffolds for Biomedical Applications

Titanium and its alloys are currently considered as one of the most important metallic materials used in the biomedical applications, due to their excellent mechanical properties and superior biocompatibility. In the present study, a new effective method for fabricating high porosity titanium alloy scaffolds was developed. Porous Ti-6Al-4V scaffolds are successfully fabricated with porosities ranging from 30% to 70% using spaceholder and powder sintering technique. Based on its acceptable properties, spherical carbamide particles with different diameters （0.56, 0.8, and 1mm） were used as the space-holder material in the present investigation. The Ti-6Al-4V scaffolds porosity is characterized by using scanning electron microscopy. The results show that the scaffolds spherical-shaped pores are depending on the shape, size and distribution of the space-holder particles. This investigation shows that the present new manufacturing technique is promising to fabricate a controlled high porosity and high purity Ti-6Al-4V scaffolds for hard tissue replacement.

Development Status and Application Analysis of Vehicle to Grid (V2G)

With the shortages of resources,environmental pollution,climate change,and other issues becoming more and more serious,it is extremely urgent to vigorously develop new energy vehicles.As the cost of batteries decrease year by year,the production and quantity of sales of electric vehicles(EVs)in the world,especially in China,increased substantially.In order to make vehicles to grid(V2G)technology better developed and applied in China.The brief introduction to V2G is given at first.Then the development status and specific cases of V2G at home and abroad are summarized.Finally,the problems that V2G may encounter during promotion and application in China are analyzed.Based on the development of the United States and Japan,specific policy recommendations are given in line with the basic national conditions of China.

Corrosion behaviour of PEEK or β-TCP-impregnated Ti6Al4V SLM structures targeting biomedical applications

Ti6Al4V cellular structures were produced by selective laser melting(SLM)and then filled either with beta-tricalcium phosphate(β-TCP)or PEEK(poly-ether-ether-ketone)through powder metallurgy techniques,to improve osteoconductivity and wear resistance.The corrosion behavior of these structures was explored considering its importance for the long-term performance of implants.Results revealed that the incorporation of open cellular pores induced higher electrochemical kinetics when being compared with dense structures.The impregnation ofβ-TCP and PEEK led to the creation of voids or gaps between the metallic matrix and the impregnated material which also influenced the corrosion behavior of the cellular structures.

Application of response surface methodology to optimize process parameters in friction welding of Ti-6Al-4V and SS304L rods

A method to decide near optimal settings of the process parameters in friction welding was proposed.The success of the friction welding process is based on various input parameters like friction pressure,friction time,upset pressure and upset time and output parameters like tensile strength,hardness and material loss.Ti-6Al-4V and SS304L(SS) materials were joined by friction welding process using interlayer techniques.The Box-Behnken design and response surface methodology(RSM) were applied to deciding the number of experiments to be performed and identify the optimum process parameters for obtaining better joint strength.The results were highly encouraging.Join strength of 523 MPa was obtained at a friction pressure of 12 N/mm^2,upset pressure of40 N/mm^2,friction time of 1.2 s and upset time of 7 s.

An Iterative Compensation Algorithm for Springback Control in Plane Deformation and Its Application

In order to solve the springback problem in sheet metal forming, the trial and error method is a widely used method in the factory, which is time-consuming and costly for its non-direction and non-quantitative. Finite element simulation is an e ective method to predict the springback of complex shape parts, but its precision is sensitive to the simulation model, particularly material model and boundary conditions. In this paper, the simple iterative method is introduced to establish the iterative compensation algorithm, and the convergence criterion of iterative parameters is put forward. In addition, the new algorithm is applied to the V-free bending and stretch-bending processes, and the convergence of curvature and bending angle is proved theoretically and verified experimentally. At the same time,the iterative compensation experiments for plane bending show that, the new method can predict the next compensaantido tnh ev atlaureg ebta cseurdv oatnu trhe ew sitphri tnhgeb earcrko ro fo fe laecshs ttehsat,n s0 o. 5 th%a ta rteh eo btatraigneet db aefntedri n2 g-3 a nitgelrea tiwoitnhs.t Thhei se rrreosre aorf clhe sps rtohpaons e±s 0 a.1%new iterative compensation algorithm to predict springback in sheet metal forming process, where each compensation value depends only on the iteration parameter di erence before and after springback for the same forming process of same material.

Mechanism, behaviour and application of iron nitrate modified carbon nanotube composites for the adsorption of arsenic in aqueous solutions

In this study,ferric nitrate modified carbon nanotube composites (FCNT) were prepared by isovolumetric impregnation using carbon nanotubes (CNTs) as the carrier and ferric nitrates the active agent.The batch experiments showed that FCNT could effectively oxidize As(III) to As(V) and react with it to form stable iron arsenate precipitates.When the dosage of FCNT was 0.1 g·L^(–1),pH value was 5–6,reaction temperature was 35℃ and reaction time was 2 h,the best arsenic removal effect could be achieved,and the removal rate of As(V) could reach 99.1%,which was always higher than 90%under acidic conditions.The adsorption results of FCNT were found to be consistent with Langmuir adsorption by static adsorption isotherm fitting,and the maximum adsorption capacity reached 118.3 mg·g^(-1).The material phase and property analysis by scanning electron microscopy,Brunauer–Emmett–Teller,Fourier transform infrared spectoscopy,X-ray photoelectron spectroscopy and other characterization methods,as well as adsorption isotherm modeling,were used to explore the adsorption mechanism of FCNT on arsenic.It was found that FCNT has microporous structure and nanostructure,and iron nanoparticles are loosely distributed on CNTs,which makes the material have good oxidation,adsorption and magnetic separation properties.Arsenic migrates on the surface of FCNT composites is mainly removed by forming insoluble compounds and co-precipitation.All the results show that FCNT treats arsenic at low cost with high adsorption efficiency,and the results also provide the experimental data basis and theoretical basis for arsenic contamination in groundwater.

New Application of SMAT in Improving Fatigue Strength of 3D Laser-printed Ti6Al4V

As a rapidly emerging additive manufacturing technique in recent years,3D laser printing technique has the significant advantages of high dimensional accuracy and high density,and can be used to produce near-net-shape metal components with complex geometric structures[1-2]. Selective laser melting (SLM) 3D printed titanium alloy has been widely used in national defense,aerospace,biomedical applications,such as aircraft structural parts,engine components,orthopaedic implants.

Image recognition and empirical application of desert plant species based on convolutional neural network

In recent years,deep convolution neural network has exhibited excellent performance in computer vision and has a far-reaching impact.Traditional plant taxonomic identification requires high expertise,which is time-consuming.Most nature reserves have problems such as incomplete species surveys,inaccurate taxonomic identification,and untimely updating of status data.Simple and accurate recognition of plant images can be achieved by applying convolutional neural network technology to explore the best network model.Taking 24 typical desert plant species that are widely distributed in the nature reserves in Xinjiang Uygur Autonomous Region of China as the research objects,this study established an image database and select the optimal network model for the image recognition of desert plant species to provide decision support for fine management in the nature reserves in Xinjiang,such as species investigation and monitoring,by using deep learning.Since desert plant species were not included in the public dataset,the images used in this study were mainly obtained through field shooting and downloaded from the Plant Photo Bank of China(PPBC).After the sorting process and statistical analysis,a total of 2331 plant images were finally collected(2071 images from field collection and 260 images from the PPBC),including 24 plant species belonging to 14 families and 22 genera.A large number of numerical experiments were also carried out to compare a series of 37 convolutional neural network models with good performance,from different perspectives,to find the optimal network model that is most suitable for the image recognition of desert plant species in Xinjiang.The results revealed 24 models with a recognition Accuracy,of greater than 70.000%.Among which,Residual Network X_8GF(RegNetX_8GF)performs the best,with Accuracy,Precision,Recall,and F1(which refers to the harmonic mean of the Precision and Recall values)values of 78.33%,77.65%,69.55%,and 71.26%,respectively.Considering the demand factors of hardware equipment and inference time,Mobile NetworkV2 achieves the best balance among the Accuracy,the number of parameters and the number of floating-point operations.The number of parameters for Mobile Network V2(MobileNetV2)is 1/16 of RegNetX_8GF,and the number of floating-point operations is 1/24.Our findings can facilitate efficient decision-making for the management of species survey,cataloging,inspection,and monitoring in the nature reserves in Xinjiang,providing a scientific basis for the protection and utilization of natural plant resources.

Stream Network Modelling from Aster GDEM Using ArcHydro GIS: Application to the Upper Moulouya River Basin (Eastern, Morocco)

This paper discusses the integration between GIS and hydrological models and presents a case study relating to the upper section of Moulouya River Basin (UMRB) situated in the east of Morocco. The Basin is an inland water-shed with a total area of approximately 10,000 km2, stretching in the junction between the Middle Atlas, the High Atlas Mountain and the Middle Moulouya basin. From ArcGIS ArcHydro framework data models, different parameters of the Moulouya River and its catchment area have been defined. DEM based ArcHydro model was run on Aster-GDEM V2 data at a horizontal spatial resolution of 30 meters. Several raster and vector products of the Upper Moulouya River and its catchment area have been defined at the end of the model. Final results of the models were discussed and compared with the reality. These results can be used in baseline for advanced hydrology and geomorphology research on the catchment area. They can support for decision-making on ground and surface water resource, distribution and management.