A data reduction and compression description for high throughput time-resolved electron microscopy

Abstract

Fast, direct electron detectors have significantly improved the spatio-temporal resolution of electron microscopy movies. Preserving both spatial and temporal resolution in extended observations, however, requires storing prohibitively large amounts of data. Here, we describe an efficient and flexible data reduction and compression scheme (ReCoDe) that retains both spatial and temporal resolution by preserving individual electron events. Running ReCoDe on a workstation we demonstrate on-the-fly reduction and compression of raw data streaming off a detector at 3 GB/s, for hours of uninterrupted data collection. The output was 100-fold smaller than the raw data and saved directly onto network-attached storage drives over a 10 GbE connection. We discuss calibration techniques that support electron detection and counting (e.g., estimate electron backscattering rates, false positive rates, and data compressibility), and novel data analysis methods enabled by ReCoDe (e.g., recalibration of data post acquisition, and accurate estimation of coincidence loss).

Real-Time Interactive 4D-STEM Phase-Contrast Imaging From Electron Event Representation Data: Less computation with the right representation

“…Until recently, the connected hardware's data-transfer capabilities limited the frame-rate of DEDs and often the size of the recorded datasets [9], [10]. The reason for this limitation is that the sensor data was transferred uncompressed over the network to data-processing hardware, such that the possible frame-rate was intrinsically limited by the data transfer rate.…”

“…The reason for this limitation is that the sensor data was transferred uncompressed over the network to data-processing hardware, such that the possible frame-rate was intrinsically limited by the data transfer rate. A straightforward way to remove this data-transfer bottleneck is to reduce and/or compress the data as they arrive in memory, and strategies to this end have been developed recently for DEDs [9], [10]. DEDs such as the detectors employed in cryo-EM and the 4D Camera are operated with experimental parameters that naturally allow data reduction and compression by only recording a limited number of electrons per frame.…”

“…Three different encoding strategies have been proposed: First, direct encoding as a linear index into the camera coordinates, second, run-length encoding of the pixel distances between electron hits, and third, binary encoding as a 2D binary array [9], [10]. The 4D Camera software currently implements linear index encoding, while the Falcon 3 and 4 cameras implement run-length encoding.…”

“…Many operations scale with the dense format linear index encoding run-length encoding bin number of counts when linear index encoding or run-length encoding are used, whereas they scale with the number of pixels when a dense data format is used. Re-coding run-length encoded data into linear-index encoded format requires O(C) operations, therefore the run-length encoded format is only slightly more inefficient for common pre-processing operations in 4D-STEM, while offering a compression advantage [10]. Given that the number of counts is 40 to 100 times lower than the number of pixels in back-thinned detectors [9], this results in a 40 to 100-fold speedup in computation time for these operations.…”

Real-time interactive ptychography from electron event representation data

“…In this talk we show with experimental data from the 4D camera that linear-index encoded EER [2,3] and its direct use in 4D-STEM phase-contrast imaging methods enables real-time, interactive phase-contrast from large-area 4D-STEM datasets.…”