A Scalable Neuromorphic Architecture to Efficiently Compute Spatial Image Filtering of High Image Resolution and Size

2020 ◽  
Vol 18 (02) ◽  
pp. 327-335
Author(s):  
Marco Abarca ◽  
Giovanny Sanchez ◽  
Luis Garcia ◽  
Juan Gerardo Avalos ◽  
Thania Frias ◽  
...  
Keyword(s):  
Image Filtering ◽  
Image Resolution ◽  
Spatial Image ◽  
High Image Resolution ◽  
Neuromorphic Architecture

A Scalable Neuromorphic Architecture to Efficiently Compute Spatial Image Filtering of High Image Resolution and Size

2019 ◽  
Vol 18 (02) ◽  
pp. 327-335
Author(s):  
Marco Abarca ◽  
Giovanny Sanchez ◽  
Luis Garcia ◽  
Juan Gerardo Avalos ◽  
Thania Frias ◽  
...  
Keyword(s):  
Image Filtering ◽  
Image Resolution ◽  
Spatial Image ◽  
High Image Resolution ◽  
Neuromorphic Architecture

Methodology for TEM Analysis of Barrier Profiles

2002 ◽  
Author(s):  
Tan-Chen Lee ◽  
Jui-Yen Huang ◽  
Li-Chien Chen ◽  
Ruey-Lian Hwang ◽  
David Su
Keyword(s):  
Sample Preparation ◽  
Profile Analysis ◽  
Image Interpretation ◽  
Sample Thickness ◽  
Image Resolution ◽  
Specimen Preparation ◽  
Tem Analysis ◽  
Preparation Methods ◽  
High Image Resolution ◽  
Common Technique

Abstract Device shrinkage has resulted in thinner barriers and smaller vias. Transmission Electron Microscopy (TEM) has become a common technique for barrier profile analysis because of its high image resolution. TEM sample preparation and image interpretation becomes difficult when the size of the small cylindrical via is close to the TEM sample thickness. Effects of different sample thickness and specimen preparation methods, therefore, have been investigated. An automatic FIB program has been shown to be useful in via sample preparation. Techniques for imaging a TEM specimen will be discussed in the paper. Conventional TEM bright field (BF) image is adequate to examine the barrieronly via; however, other techniques are more suitable for a Cu filled via.

scholarly journals Differences in the Tensor Veli Palatini Muscle and Hearing Status in Children With and Without 22q11.2 Deletion Syndrome

2019 ◽  
Vol 57 (3) ◽  
pp. 302-309
Author(s):  
Jamie L. Perry ◽  
Katelyn J. Kotlarek ◽  
Kelly Spoloric ◽  
Adriane Baylis ◽  
Lakshmi Kollara ◽  
...  
Keyword(s):  
Three Dimensional ◽  
Muscle Length ◽  
22Q11.2 Deletion Syndrome ◽  
Image Resolution ◽  
Deletion Syndrome ◽  
22Q11.2 Deletion ◽  
3 Dimensional ◽  
High Image Resolution ◽  
Tensor Veli Palatini Muscle ◽  
Magnetic Resonance Imaging Mri

Purpose: To investigate the dimensions of the tensor veli palatini (TVP) muscle using high image resolution 3-dimensional magnetic resonance imaging (MRI) of the soft palate among children with normal velopharyngeal and craniofacial anatomy and to compare values to individuals with a diagnosis of 22q11.2 deletion syndrome (22q11DS). We also sought to determine whether there is a relationship between hypoplasia of the TVP and severity of middle ear dysfunction and hearing loss. Methods: Three-dimensional MRI were used to collect and analyze data obtained across 53 children between 4 and 12 years of age, including 40 children with normal velopharyngeal and craniofacial anatomy and 13 children with a diagnosis of 22q11.2 DS. Tensor veli palatini muscle length, thickness, and volume as well as bihamular distance were compared among participant groups. Results: A Welch’s t-test revealed that the TVP in participants with 22q11DS is significantly shorter ( P = .005, 17.3 vs 19.0 mm), thinner ( P < .001, 1.1 vs 1.8 mm), and less voluminous ( P < .001, 457.5 vs 667.3 mm3) than participants without 22q11DS. Participants with 22q11DS also had a greater ( P = .006, 27.7 vs 24.7 mm) bihamular distance than participants without 22q11DS. There was an inverse relationship between TVP abnormalities noted above and the severity of audiologic and otologic histories. Conclusion: The TVP muscle is substantially reduced in volume, length, and thickness in children with 22q11DS. These findings serve as preliminary support for the association of patient hearing and otologic severity and TVP dysmorphology.

Effect of X-Ray µCT Scanning on the Growth and Activity of Microorganisms in a Heap Bioleaching System

2017 ◽  
Vol 262 ◽  
pp. 143-146
Author(s):  
Mehdi Ghadiri ◽  
Susan T.L. Harrison ◽  
Marijke A. Fagan-Endres
Keyword(s):  
Microbial Activity ◽  
Image Resolution ◽  
Cell Counting ◽  
Reaction Products ◽  
Micro Computed Tomography ◽  
X Ray ◽  
Moderately Thermophilic ◽  
Heap Bioleaching ◽  
High Image Resolution ◽  
Surface Mineral

In heap bioleaching, a process in which microorganisms are required for the regeneration of leach reagents and control of reaction products, inaccessibility of non-surface mineral grains is a key cause of low recovery and long extraction times. High resolution, non-destructive 3D X-ray micro-computed tomography (μCT) is an imaging technique that has been successfully demonstrated for the study of abiotic leaching of non-surface minerals. For this technique to be applied to biotic leaching, it is required that the iron and sulphur oxidizing abilities of the microorganisms are not affected by the irradiation experienced. In the current study, the feasibility of investigating biotic leaching by X-ray μCT is explored by examining the relative energies required to achieve the high image resolution needed for mineral grain mapping while avoiding microbial deactivation. A mixed mesophilic and moderately thermophilic culture in solution was used and exposed to various X-ray energy doses. Direct microscopic cell counting and redox potential were measured to quantify the microbial activity and growth. The results showed that exposure to X-ray does not affect microbial activity at 35-90 kV, 200-280 μA and a distance of 7.2 cm between energy source and sample, however, it has an influence at 120 and 150 kV. This indicates that while X-ray μCT does influence the microbial cultures, it can be used for bioleaching studies at lower energy doses.

scholarly journals Edge Oriented Spatial Image Filtering Technique

2011 ◽  
Vol 31 (2) ◽  
pp. 45-47
Author(s):  
Author N.Ramakrishna ◽  
Author Dr.V.S.Mallela
Keyword(s):  
Image Filtering ◽  
Spatial Image ◽  
Filtering Technique

scholarly journals Deep Neural Frameworks Improve the Accuracy of General Practitioners in the Classification of Pigmented Skin Lesions

Diagnostics ◽  
2020 ◽  
Vol 10 (11) ◽  
pp. 969
Author(s):  
Maximiliano Lucius ◽  
Jorge De All ◽  
José Antonio De All ◽  
Martín Belvisi ◽  
Luciana Radizza ◽  
...  
Keyword(s):  
General Practitioners ◽  
Skin Lesions ◽  
Image Resolution ◽  
Aggregated Data ◽  
Skin Imaging ◽  
High Image Resolution ◽  
Random Dataset ◽  
Algorithmic Framework ◽  
Diagnosis Accuracy ◽  
Pigmented Skin Lesions

This study evaluated whether deep learning frameworks trained in large datasets can help non-dermatologist physicians improve their accuracy in categorizing the seven most common pigmented skin lesions. Open-source skin images were downloaded from the International Skin Imaging Collaboration (ISIC) archive. Different deep neural networks (DNNs) (n = 8) were trained based on a random dataset constituted of 8015 images. A test set of 2003 images was used to assess the classifiers’ performance at low (300 × 224 RGB) and high (600 × 450 RGB) image resolution and aggregated data (age, sex and lesion localization). We also organized two different contests to compare the DNN performance to that of general practitioners by means of unassisted image observation. Both at low and high image resolution, the DNN framework differentiated dermatological images with appreciable performance. In all cases, the accuracy was improved when adding clinical data to the framework. Finally, the least accurate DNN outperformed general practitioners. The physician’s accuracy was statistically improved when allowed to use the output of this algorithmic framework as guidance. DNNs are proven to be high performers as skin lesion classifiers and can improve general practitioner diagnosis accuracy in a routine clinical scenario.

scholarly journals River-ice and water velocities using the Planet optical cubesat constellation

2019 ◽  
Vol 23 (10) ◽  
pp. 4233-4247 ◽  
Author(s):  
Andreas Kääb ◽  
Bas Altena ◽  
Joseph Mascaro
Keyword(s):  
Land Surface ◽  
Time Lag ◽  
Image Resolution ◽  
River Ice ◽  
Earth Surface ◽  
Spatial Image ◽  
The Amur River ◽  
Ice Floes ◽  
Image Pairs ◽  
Short Time

Abstract. The PlanetScope constellation consists of ∼150 optical cubesats that are evenly distributed like strings of pearls on two orbital planes, scanning the Earth's land surface once per day with an approximate spatial image resolution of 3 m. Subsequent cubesats on each of the orbital planes image the Earth surface with a nominal time lag of approximately 90 s between them, which produces near-simultaneous image pairs over the across-track overlaps of the cubesat swaths. We exploit this short time lag between subsequent Planet cubesat images to track river ice floes on northern rivers as indicators of water surface velocities. The method is demonstrated for a 60 km long reach of the Amur River in Siberia, and a 200 km long reach of the Yukon River in Alaska. The accuracy of the estimated horizontal surface velocities is of the order of ±0.01 m s−1. The application of our approach is complicated by cloud cover and low sun angles at high latitudes during the periods where rivers typically carry ice floes, and by the fact that the near-simultaneous swath overlaps, by design, do not cover the complete Earth surface. Still, the approach enables direct remote sensing of river surface velocities for numerous cold-region rivers at a number of locations and occasionally several times per year – which is much more frequent and over much larger areas than currently feasible. We find that freeze-up conditions seem to offer ice floes that are generally more suitable for tracking, and over longer time periods, compared with typical ice break-up conditions. The coverage of river velocities obtained could be particularly useful in combination with satellite measurements of river area, and river surface height and slope.

SEM Equipment Capabilities Evaluated for Sub-Half Micron Semiconductor Applications

1997 ◽  
Author(s):  
J. Cowan ◽  
T. Taylor
Keyword(s):  
Electron Microscopy ◽  
Memory Cell ◽  
Image Resolution ◽  
Evaluation Methodology ◽  
Reasonable Cost ◽  
High Image Resolution ◽  
High Assurance ◽  
Electron Microscopes ◽  
Scanning Electron Microscopes ◽  
Scanning Electron

Abstract Evaluation of Scanning Electron Microscopes (SEMs) was initiated for the purpose of purchasing a SEM that would improve the productivity of scanning electron microscopy during the cycle of analysis and deprocessing of semiconductor devices in a failure analysis lab. In addition to the need for high image resolution at low electron acceleration voltages, an accurate motorized stage is a major evaluation factor. It is necessary for the analyst to drive directly to a known location such as a memory cell with a high assurance that the site of interest was found. There are two main areas of focus in this paper. First, our SEM evaluation methodology will be presented along with the results of our evaluation. Second, the technology associated with motorized stages will be discussed in light of our requirements for a motorized, highly accurate stage. As a byproduct of this evaluation, this paper is presented so as to push the SEM industry to offer a SEM with an accurate stage for subhalfmicron products at reasonable cost.

Oil Palm Tree Detection from High Resolution Drone Image Using Convolutional Neural Network

2019 ◽  
Vol 1 (2) ◽  
pp. 6-9
Author(s):  
Chee Cheong Lee ◽  
See Yee Tan ◽  
Tien Sze Lim ◽  
Voon Chet Koo
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Template Matching ◽  
Window Size ◽  
Ground Truth ◽  
Frequency Domain Analysis ◽  
Image Resolution ◽  
Palm Tree ◽  
Tree Detection ◽  
High Image Resolution

We propose a method to combine several image processing methods with Convolutional Neural Network (CNN) to perform palm tree detection and counting. This paper focuses on drone imaging, which has a high image resolution and is widely deployed in the plantation industry. Analyzing drone images is challenging due to variable drone flying altitudes, resulting in inconsistent tree sizes in images captured. Counting by template matching or fixed sliding window size method often produces an inaccurate count. Instead, our method employs frequency domain analysis to estimate tree size before CNN. The method is evaluated using two images, ranging from a few thousand trees to a few hundred thousand trees per image. We have summarized the accuracy of the proposed method by comparing the results with manually labelled ground truth.

Sign in / Sign up

Export Citation Format

Share Document