Brain tumor identification is a challenging task in neuro-oncology.The brain’s complex anatomy makes it a crucial part of the central nervous system.Accurate tumor classification is crucial for clinical diagnosis and...Brain tumor identification is a challenging task in neuro-oncology.The brain’s complex anatomy makes it a crucial part of the central nervous system.Accurate tumor classification is crucial for clinical diagnosis and treatment planning.This research presents a significant advancement in the multi-classification of brain tumors.This paper proposed a novel architecture that integrates Enhanced ResNeXt 101_32×8d,a Convolutional Neural Network(CNN)with a multi-head self-attention(MHSA)mechanism.This combination harnesses the strengths of the feature extraction,feature representation by CNN,and long-range dependencies by MHSA.Magnetic Resonance Imaging(MRI)datasets were employed to check the effectiveness of the proposed architecture.The first dataset(DS-1,Msoud)included four brain tumor classes,and the second dataset(DS-2)contained seven brain tumor classes.This methodology effectively distinguished various tumor classes,achieving high accuracies of 99.75% on DS-1 and 98.80% on DS-2.These impressive results indicate the superior performance and adaptability of our model for multiclass brain tumor classification.Evaluationmetrics such as accuracy,precision,recall,F1 score,and ROC(receiver operating characteristic)curve were utilized to comprehensively evaluate model validity.The performance results showed that the model is well-suited for clinical applications,with reduced errors and high accuracy.展开更多
The liquid crystal spatial light modulator （LC SLM） is very suitable for wavefront correction and optical testing and can produce a wavefront with large phase change and high accuracy. The LC SLM is composed of thou...The liquid crystal spatial light modulator （LC SLM） is very suitable for wavefront correction and optical testing and can produce a wavefront with large phase change and high accuracy. The LC SLM is composed of thousands of pixels and the pixel size and shape have effects on the diffractive characteristics of the LC SLM. This paper investigates the pixel effect on the phase of the wavefront with the scalar diffractive theory. The results show that the maximum optical path difference modulation is 41μm to produce the paraboloid wavefront with the peak to valley accuracy better than λ/10. Effects of the mismatch between the pixel and the period, and black matrix on the diffraction efficiency of the LC SLM are also analysed with the Fresnel phase lens model. The ability of the LC SLM is discussed for optical testing and wavefront correction based on the calculated results. It shows that the LC SLM can be used as a wavefront corrector and a compensator.展开更多
Single pixel imaging is a novel imaging technique,and it becomes a focus of research in recent years due to its advantages such as high lateral resolution and high robustness to noise.Imaging speed is one of the criti...Single pixel imaging is a novel imaging technique,and it becomes a focus of research in recent years due to its advantages such as high lateral resolution and high robustness to noise.Imaging speed is one of the critical shortcomings,which limits the further development and applications of this technique.In this paper,we focus on the issues of imaging efficiency of a single pixel imaging system.We propose semi-continuous wavelet transform(SCWT)protocol and introduce the protocol into the single pixel imaging system.The proposed protocol is something between continuous wavelet transform and discrete wavelet transform,which allows the usage of those smooth(usually non-orthogonal,and they have advantages in representing smooth signals compressively,which can improve the imaging speed of single pixel imaging)wavelets and with limited numbers of measurements.The proposed imaging scheme is studied,and verified by simulations and experiments.Furthermore,a comparison between our proposed scheme and existing imaging schemes are given.According to the results,the proposed SCWT scheme is proved to be effective in reconstructing a image compressively.展开更多
Background HEPS-BPIX is a prototype of photon counting pixel detector developed for the High Energy Photon Source.It consists of 16 silicon pixel modules which should be tested individually to ensure the function and ...Background HEPS-BPIX is a prototype of photon counting pixel detector developed for the High Energy Photon Source.It consists of 16 silicon pixel modules which should be tested individually to ensure the function and performance.Purpose Due to various factors such as the non-uniformity of the processes and voltage drop,the response of each pixel in the silicon pixel module is not identical completely.The response difference of pixels can be minimized by the threshold calibration.This system is developed for the quality test and calibration of the silicon pixel modules.Methods The system consists of a mother board,a control board and a data acquisition(DAQ)system.The mother board provides necessary resources including power supplies and the fanout of calibration signals.Besides,it can be used to test the connectivity by monitoring the power states.The control board reads data out and provides the clock,trigger and configuration data for the silicon pixel module.The DAQ system sends the control commands and receives the readout data through an Ethernet link.Results Compared with the previous readout system,this designed system has a lower noise level and better scanning curves making the calibration more accurate.And it has been successfully applied to the comparison experiments of the through silicon via and wire-bonding silicon pixel modules.Conclusion The results show that this test system can be used to the quality test and calibration of the silicon pixel modules.In addition,the system can be adapted to the measurement of different pixel array detector modules.展开更多
To address the issues of small target miss detection,false positives in complex scenarios,and insufficient real-time performance in maglev train foreign object intrusion detection,this paper proposes a multi-module fu...To address the issues of small target miss detection,false positives in complex scenarios,and insufficient real-time performance in maglev train foreign object intrusion detection,this paper proposes a multi-module fusion improvement algorithm,YOLO11-FADA(Fusion of Augmented Features and Dynamic Attention),based on YOLO11.The model achieves collaborative optimization through three key modules:The Local Feature Augmentation Module(LFAM)enhances small target features and mitigates feature loss during down-sampling through multi-scale feature parallel extraction and attention fusion.The Dynamically Tuned Self-Attention(DTSA)module introduces learnable parameters to adjust attentionweights dynamically,and,in combinationwith convolution,expands the receptive field to suppress complex background interference.TheWeighted Convolution 2D(wConv2D)module optimizes convolution kernel weights using symmetric density functions and sparsification,reducing the parameter count by 30% while retaining core feature extraction capabilities.YOLO11-FADA achieves a mAP@0.5 of 0.907 on a custom maglev train foreign object dataset,improving by 3.0% over the baseline YOLO11 model.The model’s computational complexity is 7.3 GFLOPs,with a detection speed of 118.6 FPS,striking a balance between detection accuracy and real-time performance,thereby offering an efficient solution for rail transit safety monitoring.展开更多
文摘Brain tumor identification is a challenging task in neuro-oncology.The brain’s complex anatomy makes it a crucial part of the central nervous system.Accurate tumor classification is crucial for clinical diagnosis and treatment planning.This research presents a significant advancement in the multi-classification of brain tumors.This paper proposed a novel architecture that integrates Enhanced ResNeXt 101_32×8d,a Convolutional Neural Network(CNN)with a multi-head self-attention(MHSA)mechanism.This combination harnesses the strengths of the feature extraction,feature representation by CNN,and long-range dependencies by MHSA.Magnetic Resonance Imaging(MRI)datasets were employed to check the effectiveness of the proposed architecture.The first dataset(DS-1,Msoud)included four brain tumor classes,and the second dataset(DS-2)contained seven brain tumor classes.This methodology effectively distinguished various tumor classes,achieving high accuracies of 99.75% on DS-1 and 98.80% on DS-2.These impressive results indicate the superior performance and adaptability of our model for multiclass brain tumor classification.Evaluationmetrics such as accuracy,precision,recall,F1 score,and ROC(receiver operating characteristic)curve were utilized to comprehensively evaluate model validity.The performance results showed that the model is well-suited for clinical applications,with reduced errors and high accuracy.
基金Project supported by the National Natural Science Foundation of China （Nos 60578035, 50473040） and the Science Foundation of Jilin Province （Nos 20050520, 20050321-2）.
文摘The liquid crystal spatial light modulator （LC SLM） is very suitable for wavefront correction and optical testing and can produce a wavefront with large phase change and high accuracy. The LC SLM is composed of thousands of pixels and the pixel size and shape have effects on the diffractive characteristics of the LC SLM. This paper investigates the pixel effect on the phase of the wavefront with the scalar diffractive theory. The results show that the maximum optical path difference modulation is 41μm to produce the paraboloid wavefront with the peak to valley accuracy better than λ/10. Effects of the mismatch between the pixel and the period, and black matrix on the diffraction efficiency of the LC SLM are also analysed with the Fresnel phase lens model. The ability of the LC SLM is discussed for optical testing and wavefront correction based on the calculated results. It shows that the LC SLM can be used as a wavefront corrector and a compensator.
基金the Natural Science Foundation of Jilin Province,China(Grand No.YDZJ202101ZYTS030)。
文摘Single pixel imaging is a novel imaging technique,and it becomes a focus of research in recent years due to its advantages such as high lateral resolution and high robustness to noise.Imaging speed is one of the critical shortcomings,which limits the further development and applications of this technique.In this paper,we focus on the issues of imaging efficiency of a single pixel imaging system.We propose semi-continuous wavelet transform(SCWT)protocol and introduce the protocol into the single pixel imaging system.The proposed protocol is something between continuous wavelet transform and discrete wavelet transform,which allows the usage of those smooth(usually non-orthogonal,and they have advantages in representing smooth signals compressively,which can improve the imaging speed of single pixel imaging)wavelets and with limited numbers of measurements.The proposed imaging scheme is studied,and verified by simulations and experiments.Furthermore,a comparison between our proposed scheme and existing imaging schemes are given.According to the results,the proposed SCWT scheme is proved to be effective in reconstructing a image compressively.
基金the State Key Laboratoryof Particle Detection and Electronics, SKLPDE-ZZ-202002.
文摘Background HEPS-BPIX is a prototype of photon counting pixel detector developed for the High Energy Photon Source.It consists of 16 silicon pixel modules which should be tested individually to ensure the function and performance.Purpose Due to various factors such as the non-uniformity of the processes and voltage drop,the response of each pixel in the silicon pixel module is not identical completely.The response difference of pixels can be minimized by the threshold calibration.This system is developed for the quality test and calibration of the silicon pixel modules.Methods The system consists of a mother board,a control board and a data acquisition(DAQ)system.The mother board provides necessary resources including power supplies and the fanout of calibration signals.Besides,it can be used to test the connectivity by monitoring the power states.The control board reads data out and provides the clock,trigger and configuration data for the silicon pixel module.The DAQ system sends the control commands and receives the readout data through an Ethernet link.Results Compared with the previous readout system,this designed system has a lower noise level and better scanning curves making the calibration more accurate.And it has been successfully applied to the comparison experiments of the through silicon via and wire-bonding silicon pixel modules.Conclusion The results show that this test system can be used to the quality test and calibration of the silicon pixel modules.In addition,the system can be adapted to the measurement of different pixel array detector modules.
文摘To address the issues of small target miss detection,false positives in complex scenarios,and insufficient real-time performance in maglev train foreign object intrusion detection,this paper proposes a multi-module fusion improvement algorithm,YOLO11-FADA(Fusion of Augmented Features and Dynamic Attention),based on YOLO11.The model achieves collaborative optimization through three key modules:The Local Feature Augmentation Module(LFAM)enhances small target features and mitigates feature loss during down-sampling through multi-scale feature parallel extraction and attention fusion.The Dynamically Tuned Self-Attention(DTSA)module introduces learnable parameters to adjust attentionweights dynamically,and,in combinationwith convolution,expands the receptive field to suppress complex background interference.TheWeighted Convolution 2D(wConv2D)module optimizes convolution kernel weights using symmetric density functions and sparsification,reducing the parameter count by 30% while retaining core feature extraction capabilities.YOLO11-FADA achieves a mAP@0.5 of 0.907 on a custom maglev train foreign object dataset,improving by 3.0% over the baseline YOLO11 model.The model’s computational complexity is 7.3 GFLOPs,with a detection speed of 118.6 FPS,striking a balance between detection accuracy and real-time performance,thereby offering an efficient solution for rail transit safety monitoring.
