UAV Detection with Transfer Learning from Simulated Data of Laser Active Imaging

Shao Zhang; Guoqing Yang; Tao Sun; Kunyang Du; Jin Guo

doi:10.3390/app11115182

UAV Detection with Transfer Learning from Simulated Data of Laser Active Imaging

Applied Sciences ◽

10.3390/app11115182 ◽

2021 ◽

Vol 11 (11) ◽

pp. 5182

Author(s):

Shao Zhang ◽

Guoqing Yang ◽

Tao Sun ◽

Kunyang Du ◽

Jin Guo

Keyword(s):

Imaging System ◽

Simulated Data ◽

Simulation Method ◽

Previous Method ◽

Night Vision ◽

Simulated Dataset ◽

Security And Privacy ◽

Imaging Data ◽

Detection Model ◽

Active Imaging

With the development of our society, unmanned aerial vehicles (UAVs) appear more frequently in people’s daily lives, which could become a threat to public security and privacy, especially at night. At the same time, laser active imaging is an important detection method for night vision. In this paper, we implement a UAV detection model for our laser active imaging system based on deep learning and a simulated dataset that we constructed. Firstly, the model is pre-trained on the largest available dataset. Then, it is transferred to a simulated dataset to learn about the UAV features. Finally, the trained model is tested on real laser active imaging data. The experimental results show that the performance of the proposed method is greatly improved compared to the model not trained on the simulated dataset, which verifies the transferability of features learned from the simulated data, the effectiveness of the proposed simulation method, and the feasibility of our solution for UAV detection in the laser active imaging domain. Furthermore, a comparative experiment with the previous method is carried out. The results show that our model can achieve high-precision, real-time detection at 104.1 frames per second (FPS).

Download Full-text

Quantifying accuracy and heterogeneity in single-molecule super-resolution microscopy

10.1101/721837 ◽

2019 ◽

Author(s):

Hesam Mazidi ◽

Tianben Ding ◽

Arye Nehorai ◽

Matthew D. Lew

Keyword(s):

Single Molecule ◽

Computational Models ◽

Amyloid Fibrils ◽

Imaging System ◽

Simulated Data ◽

Super Resolution ◽

Ground Truth ◽

Localization Algorithm ◽

Reconstruction Algorithms ◽

Imaging Data

The resolution and accuracy of single-molecule localization micro-scopes (SMLMs) are routinely benchmarked using simulated data, calibration “rulers,” or comparisons to secondary imaging modalities. However, these methods cannot quantify the nanoscale accuracy of an arbitrary SMLM dataset. Here, we show that by computing localization stability under a well-chosen perturbation with accurate knowledge of the imaging system, we can robustly measure the confidence of individual localizations without ground-truth knowledge of the sample. We demonstrate that our method, termed Wasserstein-induced flux (WIF), measures the accuracy of various reconstruction algorithms directly on experimental 2D and 3D data of microtubules and amyloid fibrils. We further show that WIF confidences can be used to evaluate the mismatch between computational models and imaging data, enhance the accuracy and resolution of recon-structed structures, and discover hidden molecular heterogeneities. As a computational methodology, WIF is broadly applicable to any SMLM dataset, imaging system, and localization algorithm.

Download Full-text

Quantifying accuracy and heterogeneity in single-molecule super-resolution microscopy

Nature Communications ◽

10.1038/s41467-020-20056-9 ◽

2020 ◽

Vol 11 (1) ◽

Author(s):

Hesam Mazidi ◽

Tianben Ding ◽

Arye Nehorai ◽

Matthew D. Lew

Keyword(s):

Single Molecule ◽

Computational Models ◽

Amyloid Fibrils ◽

Imaging System ◽

Simulated Data ◽

Super Resolution ◽

Ground Truth ◽

Localization Algorithm ◽

Reconstruction Algorithms ◽

Imaging Data

AbstractThe resolution and accuracy of single-molecule localization microscopes (SMLMs) are routinely benchmarked using simulated data, calibration rulers, or comparisons to secondary imaging modalities. However, these methods cannot quantify the nanoscale accuracy of an arbitrary SMLM dataset. Here, we show that by computing localization stability under a well-chosen perturbation with accurate knowledge of the imaging system, we can robustly measure the confidence of individual localizations without ground-truth knowledge of the sample. We demonstrate that our method, termed Wasserstein-induced flux (WIF), measures the accuracy of various reconstruction algorithms directly on experimental 2D and 3D data of microtubules and amyloid fibrils. We further show that WIF confidences can be used to evaluate the mismatch between computational models and imaging data, enhance the accuracy and resolution of reconstructed structures, and discover hidden molecular heterogeneities. As a computational methodology, WIF is broadly applicable to any SMLM dataset, imaging system, and localization algorithm.

Download Full-text

A 3D Outdoor Scene Scanner Based on a Night-Vision Range-Gated Active Imaging System

Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06) ◽

10.1109/3dpvt.2006.8 ◽

2006 ◽

Cited By ~ 19

Author(s):

David Monnin ◽

Armin L. Schneider ◽

Frank Christnacher ◽

Yves Lutz

Keyword(s):

Imaging System ◽

Night Vision ◽

Active Imaging

Download Full-text

Semiconductor Laser Night Vision Image of Vehicle Recognition Based on Image Processing

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.347-350.947 ◽

2013 ◽

Vol 347-350 ◽

pp. 947-950

Author(s):

Zhao Xin Wang ◽

Shuang Xu ◽

Ling Gao ◽

Hui Zhang ◽

Yan Nan Zhai

Keyword(s):

Semiconductor Laser ◽

Optical System ◽

Imaging System ◽

Optical Axis ◽

Night Vision ◽

Laser Diode Array ◽

Vehicle Recognition ◽

Semiconductor Laser Diode ◽

Dsp System ◽

Active Imaging

The semiconductor laser night imaging system introduced in this paper is an active imaging system. The system using laser as light source,can get the night scene in total darkness conditions at night. So this system has been widely used in the field of public security and vehicle recognition.This system is composed of semiconductor laser diode array, laser transmitting and receiving optical system, camera receiver and DSP system and so on. The main character of this system is the technology of non-scanning laser imaging to obtain laser range images, and then the images are under the treatments of real-time collection, storing, superpose filtering, and characteristics extraction. With the program of centroid coordinate of the target, the optical axis of the optical system will be in coincidence and the action order will be formed.

Download Full-text

Correlation of Pore Structure and Permeability by SEM-Image Analysis

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100180896 ◽

1990 ◽

Vol 48 (1) ◽

pp. 428-429

Author(s):

C. A. Callender ◽

Wm. C. Dawson ◽

J. J. Funk

Keyword(s):

Image Analysis ◽

Pore Structure ◽

Imaging System ◽

Pore Space ◽

Simulation Models ◽

Image Analyzer ◽

Imaging Data ◽

Sem Images ◽

Thin Sections ◽

Rock Types

The geometric structure of pore space in some carbonate rocks can be correlated with petrophysical measurements by quantitatively analyzing binaries generated from SEM images. Reservoirs with similar porosities can have markedly different permeabilities. Image analysis identifies which characteristics of a rock are responsible for the permeability differences. Imaging data can explain unusual fluid flow patterns which, in turn, can improve production simulation models.Analytical SchemeOur sample suite consists of 30 Middle East carbonates having porosities ranging from 21 to 28% and permeabilities from 92 to 2153 md. Engineering tests reveal the lack of a consistent (predictable) relationship between porosity and permeability (Fig. 1). Finely polished thin sections were studied petrographically to determine rock texture. The studied thin sections represent four petrographically distinct carbonate rock types ranging from compacted, poorly-sorted, dolomitized, intraclastic grainstones to well-sorted, foraminiferal,ooid, peloidal grainstones. The samples were analyzed for pore structure by a Tracor Northern 5500 IPP 5B/80 image analyzer and a 80386 microprocessor-based imaging system. Between 30 and 50 SEM-generated backscattered electron images (frames) were collected per thin section. Binaries were created from the gray level that represents the pore space. Calculated values were averaged and the data analyzed to determine which geological pore structure characteristics actually affect permeability.

Download Full-text

Broadband near-infrared emission of K3ScF6:Cr3+ phosphors for night vision imaging system sources

Chemical Engineering Journal ◽

10.1016/j.cej.2021.129271 ◽

2021 ◽

Vol 417 ◽

pp. 129271

Author(s):

Haojun Yu ◽

Jian Chen ◽

Ruiyu Mi ◽

Juyu Yang ◽

Yan-gai Liu

Keyword(s):

Near Infrared ◽

Imaging System ◽

Infrared Emission ◽

Night Vision ◽

Near Infrared Emission

Download Full-text

EXPRESS: Design Considerations for Discrete Frequency Infrared Microscopy Systems

Applied Spectroscopy ◽

10.1177/00037028211013372 ◽

2021 ◽

pp. 000370282110133

Author(s):

Rohit Bhargava ◽

Yamuna Dilip Phal ◽

Kevin Yeh

Keyword(s):

Imaging System ◽

Chemical Imaging ◽

Optical Component ◽

Imaging Systems ◽

Imaging Data ◽

Discrete Frequency ◽

Figures Of Merit ◽

Design Considerations ◽

Hardware Implementations ◽

Measurement Capabilities

Discrete frequency infrared (DFIR) chemical imaging is transforming the practice of microspectroscopy by enabling a diversity of instrumentation and new measurement capabilities. While a variety of hardware implementations have been realized, considerations in the design of all-IR microscopes have not yet been compiled. Here we describe the evolution of IR microscopes, provide rationales for design choices, and the major considerations for each optical component that together comprise an imaging system. We analyze design choices in illustrative examples that use these components to optimize performance, under their particular constraints. We then summarize a framework to assess the factors that determine an instrumentâs performance mathematically. Finally, we summarize the design and analysis approach by enumerating performance figures of merit for spectroscopic imaging data that can be used to evaluate the capabilities of imaging systems or suitability for specific intended applications. Together, the presented concepts and examples should aid in understanding available instrument configurations, while guiding innovations in design of the next generation of IR chemical imaging spectrometers.

Download Full-text

A real time detection method for abnormal strapping of steel coil based on CCD active imaging

Measurement Science and Technology ◽

10.1088/1361-6501/ac3856 ◽

2021 ◽

Author(s):

Chuan Ye ◽

Liming Zhao ◽

Qiyan Wang ◽

Bo Pan ◽

Youchun Xie ◽

...

Keyword(s):

Maximum Entropy ◽

Detection Method ◽

Saliency Detection ◽

Automatic Segmentation ◽

Engineering Application ◽

Detection Accuracy ◽

Steel Coil ◽

Detection Model ◽

Laser Stripe ◽

Active Imaging

Abstract In order to accurately detect the abnormal looseness of strapping in the process of steel coil hoisting, an accurate detection method of strapping abnormality based on CCD structured light active imaging is proposed. Firstly, a maximum entropy laser stripe automatic segmentation model integrating multi-scale saliency features is constructed. With the help of saliency detection model, the purpose is to reduce the interference of the environment to the laser stripe and highlight the distinguishability between the stripe and the background. Then, the maximum entropy is used to segment the fused saliency features and accurately extract the stripe contour. Finally, the stripe normal field is obtained by calculating the stripe gradient vector, the stripe center line is extracted based on the stripe distribution normal direction, and the abnormal strapping is recognized online according to the stripe center. Experiments show that the proposed method is effective in terms of detection accuracy and time efficiency, and has certain engineering application value.

Download Full-text

The Time-Driven Activity-Based Costing Model for a Small Startup in Indonesia

Binus Business Review ◽

10.21512/bbr.v11i2.5869 ◽

2020 ◽

Vol 11 (2) ◽

pp. 141-148

Author(s):

Dewi Fitriasari ◽

Naoko Kawahara

Keyword(s):

Human Capital ◽

Development Process ◽

Simulated Data ◽

Simulation Method ◽

The Other ◽

Software Development Process ◽

Activity Based Costing ◽

Regulatory Data ◽

Costing System ◽

And Performance

The research proposed a Time-Driven Activity-Based Costing (TDABC) model for cost and human capital analysis in a small startup in Indonesia. TDABC discussed and illustrated the integration of capabilities into cost and performance in startups. The researchers applied a qualitative simulation method. The data were taken from actual regulatory data of wage and salary from the local government of Jakarta. The other data were simulated data based on a software development process for a small startup. The result indicates that the TDABC can assist the company to trace its performance and costs. Additional factors in implementing the costing system are provided for further research and practical considerations in adopting the costing system.

Download Full-text