Understanding the dynamics of nanoparticle is of key importance across broad areas of science and technology. Environmental TEM (ETEM) with direct electron detector has emerged as an indispensable technique to study dynamic processes in nanoparticle ensembles. Digitalization is the only feasible way to analyze the images at large data scale ( &gt; 2TB data). However, the noise inherent with the measurement makes it extremely difficult. Here we report a new closed-loop self-optimizing algorithm for segmentation. It utilizes the Sobel filter and Canny edge detection, further improved by triple detection and area vote. Triple detection repeats Canny edge detection three times with different parameter sets, and the results are combined into a single frame. This detection procedure adapts to different conditions such as image intensity and contrast. Area vote combines multiple frames to produce result better than individual frames. As a result of this analysis, TEM image (Figure 1a) is segmented as shown in Figure 1b, and coordinates as well as particle areas are extracted. The digitalized information provides the foundation to capture key features of particle evolution and to analyze its mechanism. For instance, the particle size distribution in terms of R (radius) & R3 (proxy of volume) is extracted (Figure 1c and 1d), together with its temporal evolution (Figure 2). We observed the crossover (at time t ~ 4 minutes) from the early stage of evolution during which the averaged particle size is increasing to the later stage when it starts decreasing (Figure 2b). In addition, individual coalescence events are detected (Figure 2c). Our segmentation package is able to digitalize noisy TEM images of supported nanoparticles with high accuracy vital for understanding the underlying physical mechanisms.

Nanoparticle growth characterization by image segmentation and particle tracking

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.