JHJie Hu

Jie Hu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jie Hu has a total of 27 co-authors. Here are the top co-authors by number of shared publications:

  • Huaiguo Xiao

    College of Engineering, Jiangxi Agricultural University, Nanchang, China

    2 shared publications
  • Le Wang

    School of Intelligent Manufacturing, Huzhou Vacational & Technical College, Huzhou 313099, People’s Republic of China

    2 shared publications
  • Haibin Hu

    College of Engineering, Jiangxi Agricultural University, Nanchang, 330045, China; Jiangxi Engineering Research Center of Animal Husbandry Facility Technology Exploitation, Nanchang, 330045, China

    1 shared publication
  • Haibin Hu

    School of Engineering, Jiangxi Agricultural University, Nanchang 330045, China

    1 shared publication
  • Hang Yu

    College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou 325035, China

    1 shared publication
  • Hong Li

    1 shared publication

Latest Publications

  • Open Access

    UMLN: Open-World Object Detection Empowered by Unsupervised Modeling and Location-Enhanced Network

    by Yangyang Huang, Jie Hu, Ronghua Luo

    Open-world object detection (OWOD) is a challenging task requiring models to detect both known and unknown objects while incrementally learning from new data. Current OWOD methods typically label regions with high objectness scores as unknown objects, relying heavily on known object supervision, leading to label bias. To address this, we propose object reconstruction error modeling, using object-level semantic information for unsupervised foreground and background modeling. Additionally, we introduce an unsupervised proposal generation method, leveraging segment anything model’s zero-shot learning to generate pseudo-labels for unknown objects. However, classifiers trained on known categories tend to bias toward them during inference. To resolve this, we propose a location-enhanced network, reframing classification as a location quality prediction task. Our method achieves a significant 37% improvement in unknown category recall (52.1%) on the Microsoft common objects in context (MS-COCO) dataset, outperforming previous state-of-the-art methods while maintaining competitive performance on known objects. Furthermore, it surpasses deformable detection transformer (DETR)-based models, achieving 10.95 frames per second, with a speed advantage over faster region-based convolutional neural network (Faster R-CNN)-based methods.


  • Deep Learning Enabled in Predicting Flight Parameters Using Piezoelectric Sensor

    by Jie Hu, Yangyang Huang

    In this article, a novel flight parameter identification system (FPIS) integrated with a piezoelectric cantilevers array, and a Transformer network for precisely predicting the flight attitude, which include the airspeed (v), angle of attack (α), and angle of sideslip (β), is proposed. A circular array configuration formed by four kinds of pizeoelectric bimorph cantilevers is designed for adapt to the aerodynamic requirements Simultaneously, several comprehensive experiments are performed to obtain the sensing characteristics of flight parameters identification and to verified the above transformer network for two commonly used conical angles in aircraft head design, including FPIS-40°and FPIS-60° prototype. The experimental results indicate that U increases with v continuously increased for both of the FPIS-40° and FPIS-60°. Moreover, comparisons of predictive and experimental results show that the absolute errors of the v, α, and β of FPIS-40° (FPIS-60°) are less than 0.02 m/s (0.01 m/s), 0.05° (0.02°), 0.04° (0.02°), respectively. Therefore, the superiority of the FPIS’s performance on predicting the flight attitude is demonstrated, and the dynamic identification accuracy of v, α, and β exceeds the previous state-of-the-art method, and the advantages are 0.01 m/s versus 0.064 m/s, 0.02° versus 0.16°, 0.02° versus 0.27°, respectively.

  • Design of Piezoelectric Tactile Sensors for Elastic Detection of Tumor Tissues

    by Jie Hu, Wenxiang He, Huaiguo Xiao

    A micro piezoelectric tactile sensor (PTS, Ø = 2.0 mm) suitable for installation on an endoscope to detect the elastic modulus of tumor tissue is designed. Simultaneously, a numerical model for PTS/biological tissue (BT) contact sensing has been established, and the sensing law of BT elastic modulus and its energy conversion law of PTS have been deeply analyzed via the Comsol. The experimental and computational results show that the ratio of the response voltage ( V1 / V2 ) between the internal and the packaging component is linearly related to the elasticity modulus of BT ( Et ) within 1.0~3.5 MPa, which is consistent with the model calculation results.


  • Open Access

    Design and Optimization of Geometry of Liquid Feed Conveyor Pipes

    by Yuwen Xia, Jie Hu, Huiyue Hu, Haibin Hu, Jiajia Xiao, Renxin Liu

    The promotion and use of liquid feeding face the challenge of insufficiently stable delivery. This issue can be resolved, in part, by using the spiral flow produced by a spiral pipe (SPP). The aim of this study is to investigate how the structural characteristics of the spiral pipe affect the flow state of the liquid feed, and for this purpose, the computational fluid dynamics (CFD) technique has been employed and the liquid feed delivery process has been simulated by means of an Eulerian two-fluid model The results reveal a significant improvement in the slurry’s homogeneity as it traveled through a spiral pipe compared with a straight pipe (STP). The swirl number normally increased with the number, length, height, and angle of the spiral pipe’s guide vanes. The solid-phase distribution was more homogeneous when values of N = 1, L = 1D, H = 3/8R, and θ = 20° were used, respectively, and the COV within 10D downstream of the outlet of the spiral pipe was 3.902% smaller than that of the straight pipe. The results of this study can be used as a reference for the design of liquid feed-conveying pipes.


  • Particle erosion in 90-Degree elbow pipe of pneumatic conveying System: Simulation and validation

    by Junshi Huang, Jianping Wen, Hong Li, Yuwen Xia, Shen Tan, Huaiguo Xiao, Wumao Duan, Jie Hu


  • Design and Optimization of Resonant Cavity Piezoelectric Cantilever Sensor for Wind Velocity and Direction Measurement

    by Jie Hu, Le Wang, Si Chen

    In this paper, a resonant cavity piezoelectric cantilever sensor (RCPCS) are designed and optimized for wind velocity and direction measurement. The RCPCS consists of a resonant cavity with a ring groove array and a free-standing piezoelectric cantilever array. The presence of the resonant cavity magnifies the local pressure at the cantilever’s boundary and also increases the cantilever’s vibration amplitude according to a new resonant cavity mathematical model. An structural optimization model is created using MATLAB and the COMSOL optimization module to increase the sensitivity of RCPCS. The analytical results demonstrate that the optimized cantilever’s strain distribution is more uniform than the original results, as well as the physical mechanism of strain and aerodynamic amplification. Several comprehensive experiments are conducted to evaluate the RCPCS prototype’s performance utilizing a new identification method of wind velocity and direction. The experimental results agree well with the mathematical solutions for the RCPCS response voltage and sensitivity, and the optimized cantilever’s sensitivity is noticeably improved over the unoptimized cantilever. Note that the comparison of experimental results demonstrates a significant reduction in the range of direction error (Δθ exp) from 0.34°~ 1.04° to 0.09°~ 0.29°. Similarly, the range of velocity error (Δν exp) is reduced significantly from 1.20%~ 2.27% to 0.14% ~ 0.80%. Additionally, the measured lower limit can reach 0.2 m/s. As a result of these findings, the superiority of the proposed RCPCS’s reaction performance is validated, and a simple, highly sensitive, and high-precision wind sensor is constructed.


  • Open Access

    Temporal and Regional Differences and Empirical Analysis on Sensitive Factors of the Corn Production Cost in China

    by Shumiao Ouyang, Jie Hu, Minli Yang, Mingyin Yao, Jinlong Lin

    The corn production cost (CPC) in China is related to national food security. However, there are few studies on the temporal and regional differences (TRD) and sensitive factors in the CPC. In this paper, the TRD of the corn production cost across various regions, as well as over the entirety of the country from 2008 to 2018, is presented. It is based on the GIS exploratory spatial data analysis method (ESDA). Simultaneously, a spatial panel model is established to conduct an empirical analysis of the main factors affecting the CPC. The results from the period in question show that the CPC in China and the three major production regions present a fluctuating growth trend, mainly associated with the increase in labor prices. Moreover, the CPC exhibits significant spatial differences, and demonstrates an overall trend of gradual increase from the east to the west. Over time, the number of relatively high-cost provinces has increased. All are located in southern mountainous and hilly corn areas. In addition, the CPCs of various regions are spatially correlated. Factors such as the scale of land management, the degree of mechanization, and socioeconomic conditions have a significantly negative impact on the CPC in China. Furthermore, the labor structure has a notably positive impact on the CPC.