A Real-Time Small Target Vehicle Detection Algorithm with an Improved YOLOv5m Network Model

To address the challenges of high complexity,poor real-time performance,and low detection rates for small target vehicles in existing vehicle object detection algorithms,this paper proposes a real-time lightweight architecture based on You Only Look Once(YOLO)v5m.Firstly,a lightweight upsampling operator called Content-Aware Reassembly of Features(CARAFE)is introduced in the feature fusion layer of the network to maximize the extraction of deep-level features for small target vehicles,reducing the missed detection rate and false detection rate.Secondly,a new prediction layer for tiny targets is added,and the feature fusion network is redesigned to enhance the detection capability for small targets.Finally,this paper applies L1 regularization to train the improved network,followed by pruning and fine-tuning operations to remove redundant channels,reducing computational and parameter complexity and enhancing the detection efficiency of the network.Training is conducted on the VisDrone2019-DET dataset.The experimental results show that the proposed algorithmreduces parameters and computation by 63.8% and 65.8%,respectively.The average detection accuracy improves by 5.15%,and the detection speed reaches 47 images per second,satisfying real-time requirements.Compared with existing approaches,including YOLOv5m and classical vehicle detection algorithms,our method achieves higher accuracy and faster speed for real-time detection of small target vehicles in edge computing.

Design of RF front end of digital BPM for BEPCII

Purpose In order to update the BPM electronic in BEPCII,a high-performance digital BPM processor is designed,and the RF front-end electronic is the key part in the digital BPM processor system.Methods According to the BEPCII physical parameters andADCunder-sampling theorem,the low-noise and high-dynamicrange RF front-end electronic module is designed.This article will introduce filter,RF amplifier,isolation,system test,etc.Results and conclusions The experimental results showed that BPM position short-term resolution is better than 100 nm,and long-term resolution is better than 5μm.It is the first domestic BPM processor that runs stably in the storage ring accelerator for a long time.Meanwhile,the study can help us to establish a solid foundation for the next work of BPM electronics development.

Automatic test system development for digital beam position monitor of HEPS and BEPCII

Purpose Hundreds of digital beam position monitor processors(DBPM)are required to be produced during the construction of projects such as High Energy Photon Source(HEPS)and the upgrade project of the Beijing Electron Positron Collider(BEPCII),which brings great challenges to the test work.In order to achieve accurate,fast,and complete mass production tests of DBPMs,an automatic test system(ATS)has been developed in this article.Methods According to the test items of DBPM,the standardized testing softwareflow is designed based on virtual instru-ment program control technology and experimental physics and industrial control system(EPICS),which realize automatic adjustment of test parameters and automatic acquisition of test result data.Results and conclusions The ATS can realize one-button testing of channel coefficients,channel linearity,attenuator linearity,beam current dependence(BCD)and sampling signal-to-noise ratio(SNR),and generate test reports.The total test time is less than 3 minutes,which is significantly more efficient compared to manual testing.More than 90 BEPCII DBPMs has been tested by this ATS in the lab.The test results proved that such a system could automatically recognize defective products and satisfy the requirements of mass testing.

Electro-mechanical offset measurements of beam position monitors

Purpose The absolute beam position measurement in the booster and storage ring of the high energy photon source(HEPS)requires the offset between the electrical and mechanical centers of the beam position monitor(BPM).Method Traditionally,an RF signal is sent to an antenna or stretched wire aligned with the vacuum chamber,and the signal at each electrode is measured.Conversely,a replacement method using a vector network analyzer,which was proposed by Lambertson,does not involve mechanical alignment and can be used for electrical offset determination.Results and conclusions The electro-mechanical offsets of booster BPMs of the HEPS were measured using the Lambertson method.The result shows good agreement with the stretched wire method within a 30μm error.

Development of new beam current transformer based on TMR

Purpose and background As the magneto-resistor technology is developed,applying it to beam instrumentation for the accelerator becomes possible.In this paper,a new principle beam current sensor based on TMR linear sensor is introduced.Methods Feasibility research by simulation and calculation has been done,a study of the critical component for the whole system has been carried out,and all these are presented in this paper.Results and conclusions Laboratory test results include the nonlinear error,and resolution analysis is introduced.Specific work plan for the next step of the improvement and upgrade is also referred to.