DDDNet: A lightweight and robust deep learning model for accurate segmentation and analysis of TEM images
Source abstract
Abstract
The primary aim of this study was to develop an optimal, lightweight model for the segmentation of transmission electron microscopy (TEM) images. Our model is designed with a minimal parameter count, superior performance metrics, and robust adaptability to variations in substrates, nanoparticle sizes, and nanomaterial diversity within TEM images. In achieving this, we benchmarked our model against four deep learning models using subsets from the Bright-Field TEM(BF-TEM) and Au-TEM datasets. Our model demonstrated exceptional segmentation performance, requiring only 0.34 M parameters and 39.33 G floating-point operations. It also provided the most accurate estimates of average nanoparticle sizes, closely matching true labeled values. These results confirm the model’s proficiency and precision in TEM image processing and introduce a powerful tool for nanoscale image analysis. Our work sets a new standard for lightweight and efficient TEM segmentation models, paving the way for future advancements in nanotechnology research.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Non-Destructive Testing Techniques · Cell Image Analysis Techniques · Image Processing Techniques and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
1 Jiangxi Province Key Laboratory of Multidimensional Intelligent Perception and Control, Jiangxi University of Science and Technology, Ganzhou 341000, China
Ascend Elements 2 , Atlanta Georgia,USA
Funding
Key Prpject of Nature Science Foundation of Jiangxi Province of china · National Natural Science Foundation of China · Training Plan for Academic and Technical Leaders of Major Disciplines in Jiangxi Province of china · Jiangxi Province Key Laboratory of Multidimensional Intelligent Perception and Control of china
References (24)
Show source-provided references
1. Nanoscience, nanotechnology and spectrometry10.1016/j.sab.2013.04.008 ↗
2. Micromolding in capillaries: Applications in materials science10.1021/ja960151v ↗
3. A critical review of orientation microscopy in SEM and TEM10.1002/crat.201100125 ↗
4. Automatic computation of the area irradiated by ultrashort laser pulses in Sb materials through texture segmentation of TEM images10.1016/s0304-3991(96)00080-0 ↗
5. Automated segmentation of nanoparticles in BF TEM images by U-Net binarization and branch and bound
6. Handcrafted vs. non-handcrafted features for computer vision classification10.1016/j.patcog.2017.05.025 ↗
7. Comparison of common segmentation techniques applied to transmission electron microscopy images10.1557/proc-0982-kk07-04 ↗
8. Multi-scale and first derivative analysis for edge detection in TEM images
9. Nanostructure quantification of hard carbon electrodes through advanced HRTEM image analysis10.1149/1945-7111/ac8e36 ↗
10. Nanostructure quantification of turbostratic carbon by HRTEM image analysis: State of the art, biases, sensitivity and best practices10.1016/j.carbon.2021.03.043 ↗
11. Development of an HRTEM image analysis method to quantify carbon nanostructure10.1016/j.combustflame.2011.01.009 ↗
12. Texture, structure and chemistry of a boron nitride fibre studied by high resolution and analytical TEM10.1016/s0955-2219(02)00002-x ↗
13. Autoadaptive algorithm for the stacking-level estimation of membranes in TEM images10.5402/2011/650546 ↗
14. A novel framework for segmentation of secretory granules in electron micrographs10.1016/j.media.2013.12.008 ↗
15. Bayesian image segmentation of transmission electron tomography 3D reconstructions
16. A hybrid CNN-random forest algorithm for bacterial spore segmentation and classification in TEM images10.1038/s41598-023-44212-5 ↗
17. Deep learning segmentation of complex features in atomic-resolution phase-contrast transmission electron microscopy images10.1017/s1431927621000167 ↗
18. Deep neural networks segment neuronal membranes in electron microscopy images
19. U-net: Convolutional networks for biomedical image segmentation
20. Deep learning for pyrolysis reactor monitoring: From thermal imaging toward smart monitoring system10.1002/aic.16452 ↗
21. H-DenseUNet: Hybrid densely connected UNet for liver and tumor segmentation from CT volumes10.1109/tmi.2018.2845918 ↗
22. Online monitoring of iron ore pellet size distribution using lightweight convolutional neural network10.1109/tase.2023.3253574 ↗
23. Generalization across experimental parameters in machine learning analysis of high resolution transmission electron microscopy datasets10.1093/micmic/ozae001 ↗
24. Automatic segmentation of inorganic nanoparticles in BF TEM micrographs10.1016/j.ultramic.2018.06.002 ↗