This conference provides a technical forum for members of both industry and academia to present their latest applications of machine learning. Machine learning has been applied to a broad domain of image/vision systems from medical imaging to consumer cameras. Learned tasks such as image recognition, noise reduction, or natural language processing, are currently being applied in many common devices such as mobile phones. Training datasets and training methods for machine learning are critical for the success of a system. Studies demonstrating the deployment and benchmarking of machine learning algorithms on specialized computer hardware is highly valuable to many groups in this field. Sensor hardware design or selection as it pertains to machine learning tasks; for example, an analysis of different camera designs and how each pertains to the performance of an image recognition task such as object detection, is of interest. Analysis of full systems that include the sensor technology, data processing hardware, and results are welcome as each area is critical for the successful application of machine learning.

Papers or tutorials reviewing the topics covered by this section are welcome. All abstracts will be reviewed by the program committee for originality and merit. Topics of interest include, but are not limited to, the following:

Algorithms Consumer Application Industrial Application Security Medicine Big Data Hardware Big Experimental Facilities ;
Conference 11843

Applications of Machine Learning 2021

1 - 5 August 2021
Digital Forum: On-demand starting 1 August
View Session ∨
  • Signal, Image, and Data Processing Plenary Session
  • Machine Vision and Manufacturing
  • Remote Sensing
  • Algorithms and Imaging
  • Big Data, Simulations, and Physics
  • Consumer Applications and E-Commerce
  • Poster Session
Signal, Image, and Data Processing Plenary Session
11841-501
Author(s): Nibir K. Dhar, U.S. Army CCDC C5ISR Center Night Vision and Electronics Sensors Directorate (United States)
Show Abstract + Hide Abstract
The warfighter’s effectiveness in current and future combat missions can be severely limited by a lack of adequate situational awareness (SA). Better target discrimination, better view of the operational scene with larger fields of regard and longer standoff distances are some of the important criteria. SA also strongly depends on the information, signal and data processing that can provide visual and analytics at the edge. This presentation will highlight current and future challenges as well as discuss a path forward to leverage the AI/ML and other imaging technologies. In addition, highlights of the new innovation platform will be presented.
Machine Vision and Manufacturing
11843-1
Author(s): Nathan Mundhenk, Ian Palmer, Brian J. Gallagher, T. Yong Han, Lawrence Livermore National Lab. (United States)
Show Abstract + Hide Abstract
We recently developed a deep learning method that can determine the critical peak stress of a material by looking at scanning electron microscope (SEM) images of the material’s crystals. However, it has been somewhat unclear what kind of image features the network is keying off of when it makes its prediction. It is common in computer vision to employ an explainable AI saliency map to tell one what parts of an image are important to the network’s decision. One can usually deduce the important features by looking at these salient locations. However, SEM images of crystals are more abstract to the human observer than natural image photographs. As a result, it is not easy to tell what features are important at the locations which are most salient. To solve this, we developed a method that helps us map features from important locations in SEM images to non-abstract textures that are easier to interpret.
11843-2
Author(s): Kyle S. Hickmann, Deborah Shutt, Vince Chiravalle, Nga Nguyen-Fotiadis, Los Alamos National Lab. (United States)
Show Abstract + Hide Abstract
In this work we demonstrate a method for leveraging high-fidelity, multi-physics simulations of high-speed impacts in a particular manufactured material to encode prior information regarding the materials strength properties. Our simulations involve a material composed of stacked cylindrical ligaments impacted by a high-velocity aluminum plate. We show that deep neural networks of relatively simple architecture can be trained on the simulations to make highly-accurate inferences of the strength properties of the material ligaments. We detail our neural architectures and the considerations that went into their design. In addition, we discuss the simplicity of our network architecture which lends itself to interpretability through derivative based sensitivity analysis.
11843-3
Author(s): Eric Bianchi, Virginia Polytechnic Institute and State Univ. (United States)
Show Abstract + Hide Abstract
Synthetic image generation using deep learning techniques like generative adversarial networks (GANS), has drastically improved since its inception. There has been significant research using human face datasets like FFHQ, and city-semantic datasets for self-driving car applications. Utilizing latent space distributions, researchers have been able to generate or edit images to exhibit specific traits in the resultant images, like face-aging. However, there has been little GAN research and datasets in the structural infrastructure domain. We propose several GAN applications for predicting infrastructure deterioration, and incrementally aging or renewing infrastructure. The prediction of deterioration over time is valuable to inspectors and engineers because it gives a window into the future on where potential defects may occur. The novel dataset used was procured from extracting hundreds of thousands of images from Virginia Department of Transportation VDOT bridge inspection reports.
11843-4
Author(s): Aneek E. James, Alexander Wang, Songli Wang, Keren Bergman, Columbia Univ. (United States)
Remote Sensing
11843-5
Author(s): Damián P. San Román Alerigi, Saudi Aramco (Saudi Arabia)
Show Abstract + Hide Abstract
We investigate the use of machine learning and hyperspectral imaging to infer multiphysics properties of materials. The work explores the use of complex neural network models to enable real-time mappings between electromagnetic and other physical properties of matter. These techniques can be deployed in-situ without requiring sample extraction, thus reducing test time and complexity. In this work, we will discuss the use of machine-learning algorithms to predict non-electromagnetic properties, including mechanical, chemical, and thermal. The methods could be integrated to characterize multiphysics properties of matter in diverse environments and across different scales. Therefore, our work has important implications to real-time and in-situ characterization of subterranean and terrestrial environments.
11843-6
Author(s): Alice M. Durieux, Matthew T. Calef, Jason Schatz, Rick Chartrand, Michael S. Warren, Descartes Labs, Inc. (United States)
Show Abstract + Hide Abstract
Leveraging SAR to detect deforestation in the tropics is crucial given its ability to operate regardless of cloud cover. However SAR based deforestation detections are prone to false positives, especially in hilly terrain and in areas subject to heavy rains. In this study we developed a post-processing approach to improve detection precision by training a machine learning model to differentiate between real deforestation and erroneous detections. The model significantly reduced detection volume when deployed over a large area. This approach shows promise to make deforestation detections more accurate and actionable.
11843-7
Author(s): Adel Asadi, Tufts Univ. (United States); Snehamoy Chatterjee, Michigan Technological Univ. (United States); Ali Imanian, Shiraz Univ. (Iran, Islamic Republic of)
Show Abstract + Hide Abstract
Landsat-7 satellite imagery lacks spatial continuity due to the failure of its Scan Line Corrector (SLC). A few methods have been proposed to fill the SLC-off imagery gaps; however, the complexity of the image patterns necessitates an efficient approach capable of reconstructing heterogeneous areas of interest. This study presents a novel pixel-based multiple-point geostatistical (MPS) method for stochastic gap-filling of Landsat-7 imagery using the spatial multi-band patterns extracted from the non-gap regions. The proposed approach was tested in areas with different land cover classes, and satisfactory qualitative and quantitative accuracy results were achieved.
11843-8
Author(s): Ryan Connal, Wade Pines, Meg Borek, Timothy Bauch, Nina Raqueno, Rochester Institute of Technology (United States); Brian d'Entremont, Alfred Garrett, Savannah River National Lab. (United States); Carl Salvaggio, Rochester Institute of Technology (United States)
Show Abstract + Hide Abstract
The process of automatically segmenting objects from complex backgrounds is extremely useful when trying to utilize those objects for computer vision research, such as object detection, autonomous driving, pedestrian tracking, etc. Instance segmentation is imperative towards ongoing research between the Digital Imaging and Remote Sensing Laboratory at the Rochester Institute of Technology and the Savannah River National Laboratory. This research involves careful experimental design towards the capture and development of a custom dataset to utilize instance segmentation on this custom dataset with the Matterport Mask R-CNN implementation [1], where condensed water vapor plumes were identified and masked from various complex backgrounds for the purpose of 3D reconstruction. [1] W. Abdulla. ”Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow.” Matterport, 2017. https://github.com/matterport/Mask RCNN
11843-9
Author(s): Christopher X. Ren, Michal Kucer, Los Alamos National Lab. (United States)
Show Abstract + Hide Abstract
The launch of global coverage satellites such as Sentinel-2 and Landsat-8 has led to an extremely large and rich volume of remote-sensing data available for analysis. Automated tools designed to manage, search and retrieve images are crucial for the exploitation of this data. It has been shown that deep learning is particularly effective for image retrieval in a remote-sensing setting. In this work, we demonstrate how neural networks can be used for remote-sensing image based retrieval on several benchmark datasets. We then perform adversarial attacks on these trained networks, in an effort to simulate a potential disruptive attack on our models with the goal of causing them to retrieve misclassify, or “misretrieve”, images based on an input image. Finally, we investigate how to "robustify" our networks against adversarial attacks.
Show Abstract + Hide Abstract
The main contribution of this work is using switched activation functions for classification and feature selection on forest images. The fast and less accurate activation function is used as long as the derivatives of the dynamical neurons are more than a given threshold and switching to the slow and more accurate activation function when the derivative gets lower than the threshold. For this purpose, memory artificial neural networks are needed. Finally, the following question is answered: if the switched activation functions provide fast and accurate results at the same time on feature selection and classification on forest images.
11843-12
Author(s): Amir Shirkhodaie, Branndon Jones, Ali Ahmadibeni, Tennessee State Univ. (United States)
Show Abstract + Hide Abstract
In this study, we used IRIS electromagnetic modeling and simulation (IRIS-EM) virtual environment system to systematically generate a large-scale synthetic SAR imagery dataset of the test marine scenarios. A typical marine scenario includes CAD models of test marine vehicle(s) and their associated wake(s) as well as the ocean layer. Prior to generating the synthetic images of each test scenario, we augment to each CAD model in the virtual environment appropriate physics properties. In this paper, we present our systematic approach for generating synthetic SAR Imagery of marine test scenarios and detail our methodology for annotating them properly. To evaluate and verify the effectiveness of this approach, we bench-marked our generated simulated marine SAR imagery with similar context images taken by the physical SAR imaging systems.
11843-13
Author(s): Amir Shirkhodaie, Ali Ahmadibeni, Branndon Jones, Tennessee State Univ. (United States)
Show Abstract + Hide Abstract
In this study, we considered three ways to construct our SAR training imagery including: (1) SAR-GAN dataset generated based on different variants of SAR imagery examples from IRIS-SAR and MSTAR datasets; (2) IRIS-SAR synthetic imagery datasets generated based on IRIS Electromagnetic physics-based model; and (3) a mixture of SAR-GAN and IRIS-SAR datasets. This paper, discusses the training results of our classifier models and describes the effectiveness of this method. Furthermore, we examine the effects of targeted and non-targeted adversarial attacks on the classifier using three techniques and test the trained classifier models. Lastly, we compare the quality and effectiveness proposed methods and discuss the aspects of development of the SAR-GAN-CNN model and present our future research contributions.
Algorithms and Imaging
11843-14
Author(s): Michal Kucer, Diane Oyen, Los Alamos National Lab. (United States)
11843-15
Author(s): Rajiv Mandya Nagaraju, Cory Merkel, Rochester Institute of Technology (United States)
Show Abstract + Hide Abstract
Producing a better segmentation is crucial in scene understanding. Semantic Segmentation is a vital task for applications such as autonomous driving, robotics, medical image understanding. Efficient high and low-level context manipulation is a key for competent pixel-level classification. In our implementation, We use a two bridged network. The first bridge manipulates the subtle differences between images and produces the mutual gate vector to understand the low-level features in the images better. The second bridge uses the dilated convolutional network to avoid attrition of the size of the image while gathering a better understanding of the image's high-level features. The initial experiments have yielded an initial mean IoU of 70.1% and pixel accuracy of 94.4% on the cityscapes dataset and 34.6% on the ADE20K dataset.
11843-17
Author(s): Charles S. Strauss, Los Alamos National Lab., (United States), New Mexico Consortium (United States); Garrett T. Kenyon, Los Alamos National Lab. (United States)
Show Abstract + Hide Abstract
Deep learning has gained traction in the medical field, assisting pathologists with many important, life-changing, tasks. One of these tasks involves tumor detection. Many others have demonstrated extremely successful tumor detection neural networks, however, almost none attempt any sort of robustness analysis. In addition to developing two tumor detection neural networks, we intentionally craft adversarial examples to both and analyze the perturbations used in each attack. We demonstrate successful imperceptible adversarial examples to a deep learning based tumor detection model, and a sparse coding based tumor detection model. Because sparse coding has previously proved to be more impervious to transferrable adversarial examples targeting deep learning based models, we developed a novel approach for producing adversarial images to sparse coding based models.
11843-19
Author(s): Marian Anghel, Cristina Garcia-Cardona, Patrick Kelly, Nicolas Hengartner, Los Alamos National Lab. (United States)
Show Abstract + Hide Abstract
Quantifying the predictive uncertainty of Neural Network (NN) models remains a dificult, unsolved problem especially since the ground truth is usually not available. In this work we evaluate many regression uncertainty estimation models and discuss their accuracy using training sets where the uncertainty is known exactly. We compare three regression models, a homoscedastic model, a heteroscedastic model, and a quantile model and show that: while all models can learn an accurate estimation of response, the accurate estimation of uncertainty is very difficult; the quantile model has the best performance in estimating uncertainty; model bias is confused with uncertainty and it is very difficult to disentangle the two when we have only one measurement per training point; improved accuracy of the estimated uncertainty is possible, but the experimental cost for learning uncertainty is very large since it requires multiple estimations of the response almost everywhere in the input space.
11843-20
Author(s): Matthias Hermann, Hochschule Konstanz (Germany)
Show Abstract + Hide Abstract
Parametric texture models have been applied successfully to synthesize artificial images. Psychophysical studies show that under defined conditions observers are unable to differentiate between model-generated and original natural textures. In industrial applications the reverse case is of interest: a texture analysis system should decide if human observers are able to discriminate between a reference and a novel texture. Here, we implemented a human-vision-inspired novelty detection approach. Assuming that the features used for texture synthesis are important for human texture perception, we compare psychophysical as well as learnt texture representations based on activations of a pretrained CNN in a novelty detection scenario. Based on a digital print inspection scenario we show that psychophysical texture representations are able to outperform CNN-encoded features.
11843-21
Author(s): Reid B. Porter, Los Alamos National Lab. (United States); Beate G. Zimmer, Texas A&M Univ. Corpus Christi (United States)
Show Abstract + Hide Abstract
Ordered Hypothesis Machines (OHM) are a class of multiplier free neural network that have the potential for extremely low power implementation on dedicated digital devices. In this paper, we describe OHM neural networks and how they relate to a number of other neuromorphic architectures. We outline a most significant bit (MSB) first, bit-serial implementation for OHM networks that enables static and dynamic optimizations of per sample word length without any loss in precision. We describe algorithms for gradient descent in software as well as algorithms for mapping trained OHM networks to the MSB-first bit-serial architecture. Our experiments estimate the computational efficiency of the proposed approach for networks trained on benchmark data.
11843-22
Author(s): Egor Sedov, Novosibirsk State Univ. (Russian Federation), Aston Institute of Photonic Technologies, Aston Univ. (United Kingdom); Yaroslav Prylepskiy, Aston Institute of Photonic Technologies, Aston Univ. (United Kingdom); Igor Chekhovskoy, Novosibirsk State Univ. (Russian Federation); Sergei Turitsyn, Aston Institute of Photonic Technologies, Aston Univ. (United Kingdom), Novosibirsk State Univ. (Russian Federation)
Show Abstract + Hide Abstract
In this work we demonstrate that the high-accuracy computation of the continuous nonlinear spectrum can be performed by using artificial neural networks. We propose the artificial neural network (NN) architecture that can efficiently perform the nonlinear Fourier (NF) optical signal processing. The NN consists of sequential convolution layers and fully connected output layers. This NN predicts only one component of the continuous NF spectrum, such that two identical NNs have to be used to predict the real and imaginary parts of the reflection coefficient. To train the NN, we precomputed 94035 optical signals. 9403 signals were used for validation and excluded from training. The final value of the relative error for the entire validation dataset was less than 0.3%. Our findings highlight the fundamental possibility of using the NNs to analyze and process complex optical signals, when the conventional algorithms can fail to deliver an acceptable result.
Big Data, Simulations, and Physics
11843-23
Author(s): Daniel Wang, New Mexico Consortium (United States), Los Alamos National Lab. (United States); Howard Pritchard, New Mexico Consortium (United States); Garrett T. Kenyon, Los Alamos National Lab. (United States)
Show Abstract + Hide Abstract
Effective up-sampling based on convolutional sparse coding may offer an efficient method for generating high quality, high resolution computational fluid dynamics simulations from lower quality, lower resolution simulations. Sparse coding, which incorporates lateral inhibition, implements neural network architecture that are more biologically accurate than classic deep learning algorithms. Sparse coding neural networks can up-sample and extrapolate spatiotemporally high resolution frames from a low-resolution or decimated input. The quality of the reconstruction can be compared against both an original non-decimated input and the reconstruction quality of competing deep learning networks.
11843-24
Author(s): Nga Nguyen-Fotiadis, Los Alamos National Lab. (United States); Garry R. Maskaly, Lawrence Livermore National Lab. (United States); Andy S. Liao, Christopher L. Fryer, John L. Kline, Kyle S. Hickmann, Los Alamos National Lab. (United States)
Show Abstract + Hide Abstract
We demonstrate and analyze a new methodology for detecting and characterizing shocks in high energy density physics (HEDP) experiments using observed radiographs. Our method consists of simulating many variations of an HEDP experiment using a multi-physics modeling code to produce synthetic radiographs that emulate the actual experimental data collection setup. Shock contours are defined by the peaks of the density derivative values obtained at each radial coordinate of an X-rayed cylindrical object, giving us a ground truth of shock position that would not be available directly from the observed radiograph. We investigate four different state-of-the-art deep convolutional neural networks, Xception, ResNet152, VGG19, and U-Net, for use in regressing the HEDP radiograph to the shock position. We find that the different network architectures are better tuned for locating distinct shock characteristics, equivalent to detecting shock-waves at multiple scales.
11843-25
Author(s): Adam Good, Lissa Moore, Garrett T. Kenyon, Howard Pritchard, Los Alamos National Lab. (United States)
11843-26
Author(s): Aaron Dant, Steve Kacenjar, Rukan Shao, Ron Neely, ASRC Federal Holding, Co. (United States)
Show Abstract + Hide Abstract
Synthetic generation of images provides a viable approach to training machine learning models when real data is sparse.  However, the process of rendering useful synthetic training data takes significant time. This paper examines the utility of a self-directed feedback training method that improves the convergence rate when training such models by using small batches of training data and observing the classification performance for each class being simulated. The performance results determine adjustments to subsequent training cases. Training of a person re-identification model with and without this feedback control is assessed.
11843-27
Author(s): Haimabati Dey, Peter Bermel, Purdue Univ. (United States)
11843-28
Author(s): Naveenta Gautam, Indian Institute of Technology, Delhi (India); Amol Choudhary, Brejesh Lall, Indian Institute of Technology Delhi (India)
Consumer Applications and E-Commerce
11843-29
Author(s): Daniel Adams, Cory Merkel, Rochester Institute of Technology (United States)
Show Abstract + Hide Abstract
Smart assistant usage has increased significantly with the AI boom and growth of IoT, though many smart assistants underperform when tasked with interpreting atypical speech input. Dysarthria, heavy accents, and Deaf and hard-of-hearing speech characteristics prove difficult for smart assistants to interpret despite the large amounts of diverse data used to train automatic speech recognition models. We explore the Transformer architecture as an automatic speech recognition model for speech with medium to low intelligibility scores. We utilize a Transformer model pre-trained on the Librispeech dataset and fine-tuned on the Torgo database. We also develop a method for utilizing proprietary automatic speech recognition models prior to interpretation on smart assistants utilizing a Raspberry Pi 4 and a Google Home smart assistant device. Initial model performance shows a 20.2% character error rate on a subset of medium intelligibility Deaf speech audio samples.
11843-30
Author(s): Siyun Wang, Karl Ni, Etsy, Inc. (United States)
11843-31
Author(s): Gaurav Anand, Siyun Wang, Karl Ni, Etsy, Inc. (United States)
11843-32
Author(s): Chris Xu, Raphael Louca, Karl Ni, Etsy, Inc. (United States)
11843-33
Author(s): Jingyuan Zhou, Patrick Callier, Murium Iqbal, Karl Ni, Etsy, Inc. (United States)
Show Abstract + Hide Abstract
To address the complexity of content personalization at Etsy's marketplace, our proposed approach aims at learning visual representations of products and directly optimizing user interactions. By leveraging metadata from our existing recommendations platform, we train image representations to predict relevant items. Such an approach is more flexible than existing baselines that use manually defined taxonomic categorizations. We use a pairwise ranking framework by adopting the Bayesian Personalized Ranking paradigm while building off of the CuratorNet approach of sampling positive and negative examples from the listing page. Additionally, we adopt an adaptive sampling approach that separately draws from both impressions and the listing population distribution at large. We discuss these contributions and their implications for an online system that is evaluated on the basis of generated revenue.
Poster Session
11843-34
Author(s): Salman Khan, Rochester Institute of Technology (United States); Anna Lang, Zylient, Inc. (United States); Carl Salvaggio, Rochester Institute of Technology (United States)
Show Abstract + Hide Abstract
Every year, 60,000 lives are lost worldwide from disasters. Building collapse during earthquakes account for the majority of these deaths. Unreinforced masonry (URM) buildings are particularly vulnerable during seismic events due to the brittle nature of the construction material. Many communities have undertaken costly and timely mitigation programs to locate and retrofit or replace them before disaster strikes. An automated approach for identifying seismically vulnerable buildings using street level imagery has been met with limited success to this point with no promising results presented in literature. We achieved the best overall accuracy reported to date, at 83.6%, in identifying unfinished URM, finished URM, and non-URM buildings. Moreover, an accuracy of 98.8% was achieved for identifying both suspected URMs (finished or unfinished URM). We perform extensive empirical analysis to establish synergistic parameters on our deep neural network, namely ResNeXt-101- FixRes.
11843-35
Author(s): Wenli Huang, Kyle King, U.S. Military Academy (United States)
Show Abstract + Hide Abstract
This paper presents the algorithms in detecting trace chemicals using a multi-wavelength camera. Multispectral images of the chemical and the background were collected using the Ocean Thin Films SpectroCam. The camera had an integrated motor with 8 filter color wheels and 8 interchangeable custom band pass filters in the spectral range of 200–900 nm. Since chemicals has its unique spectral reflectance, the stack of 8-dimensional image data was obtained and subsequently analyzed to develop an algorithm that can uniquely identify the area where a chemical is present. In this study, we primarily used RDX, 1,3,5-Trinitroperhydro-1,3,5-triazine, the explosive component in C4. The aim of this study was to investigate the potential of the multispectral imaging system and the accuracy of the model in determining C4 chemical.
11843-37
Author(s): Xin Zhang, Yi-Yung Chen, Jong-Woei Whang, Chih-Hsien Tsai, Wei-Chieh Tseng, National Taiwan Univ. of Science and Technology (Taiwan)
Show Abstract + Hide Abstract
This paper proposes a program frame that has a machine, artificial intelligence (AI), and firmware. The device is like a camera with an Infra Red (IR) and Light Emitting Diode (LED). The AI model based on Convolution Neural Network (CNN) is used to obtain the clinical pupil’s size. The CNN model uses a dilation convolution to expand the field of view (FOV) to receive a more extensive range of features and find the major and minor axes of the ellipse faster and accurately. Finally, We use an algorithm combined with firmware to measurements the patient’s eyes in real-time and obtain information about pupil size.
11843-38
Author(s): Jose Luis Haddad, Pontificia Univ. Católica de Chile (Chile), Jet Propulsion Lab. (Chile); Eduardo Bendek, Jet Propulsion Lab. (United States); Catalina Flores, Univ. Andrés Bello (Chile), Millennium Institute of Astrophysics (Chile); Umaa Rebbapragada, Mark Wronkiewicz, Jet Propulsion Lab. (United States)
11843-40
Author(s): Zahra Sobhaninia, Isfahan Univ. of Technology (Iran, Islamic Republic of); Hajar Danesh, Islamic Azad Univ. (Iran, Islamic Republic of); Rahele Kafieh, Isfahan Univ. of Medical Sciences (Iran, Islamic Republic of); Jayaram Jothi Balaji, Sankara Nethralaya (India), Medical Research Foundation (India); Vasudevan Lakshminarayanan, Univ. of Waterloo (Canada)
Show Abstract + Hide Abstract
We have developed a novel location-aware deep learning method for segmentation of the foveal avascular zone (FAZ) in en face optical coherence tomography angiography (OCTA) images. Using this we compare the FAZ related parameters determined using 4 methods, including ground truth (clinician marked image), and two other automated methods.. Our preliminary results suggest that the proposed method is a good substitute for device in-built software in and can replace the manual segmentation.
11843-41
Author(s): Aditya Chandra Mandal, Indian Institute of Technology (BHU), Varanasi (India); Jayaram Jothi balaji, Sankara Nethralaya (India), Medical Research Foundation (India); Vasudevan Lakshminarayanan, Univ. of Waterloo (Canada)
Show Abstract + Hide Abstract
. Here we present a novel approach to pupillometry that uses deep-learning (DL) methodologies applied to visible light images. We report the accuracy of our models and compare it with real-world experimental data. This work is the first step toward a visible light smartphone-based pupillometer that is fast, accurate, and relies on on-device computing
11843-42
Author(s): Enzo Casamassima, Andrew Herbert, Cory Merkel, Rochester Institute of Technology (United States)
Show Abstract + Hide Abstract
Recent studies in the field of adversarial robustness has showed that Convolutional Neural Networks (CNNs) are not as resilient as we might have though. Previous work has largely been focused on small image perturbations and how these completely throw off the classifier output, while leaving the image visually unchanged for a person. These attacks are carefully calculated at the pixel level and are unlikely to exist in a natural environment. More recent work has demonstrated that CNNs are also vulnerable to simple transformations to the input image, such as rotations and translations. These more ‘natural’ transformations are plausible to happen, either accidentally or intentionally, in a real-world scenario for a deployed system. In contrast, people are not fooled by transformations as simple as these. In this paper we explore transformations that are visually recognizable to humans, measuring the impact on the network classification accuracy.
11843-43
Author(s): Arturo Villegas, Juan P. Torres, ICFO - Institut de Ciències Fotòniques (Spain); Mario A. Quiroz-Juarez, Roberto de J. Leon-Montiel, Alfred B. U'Ren, Univ. Nacional Autónoma de México (Mexico)
Show Abstract + Hide Abstract
We demonstrate a smart laser-diffraction analysis technique for particle mixture identification. We retrieve information about the size, geometry, and ratio concentration of two-component heterogeneous particle mixtures with an efficiency above 92%. In contrast to commonly-used laser diffraction schemes—in which a large number of detectors is needed—our machine-learning-assisted protocol makes use of a single far-field diffraction pattern, contained within a small angle (∼ 0.26º) around the light propagation axis. Because of its reliability and ease of implementation, our work may pave the way towards the development of novel smart identification technologies for sample classification and particle contamination monitoring in industrial manufacturing processes.
11843-44
Author(s): Adriano da Silva Ferreira, Avaltech Innovation for Livestock (Brazil), Univ. Estadual de Campinas (Brazil); Mateus Modesto, Fabiano Rodrigues da Cunha Araujo, Avaltech Innovation for Livestock (Brazil); Roberto Daniel Sainz Gonzalez, The Univ. of California (United States), Avaltech Innovation for Livestock (Brazil)
Show Abstract + Hide Abstract
We modeled Convolutional Neural Networks (CNNs) for estimating backfat and rump fat thicknesses carcass traits from depth camera images of live cattle. We aim at designing accurate CNN models that quantify such carcass characteristics currently evaluated by ultrasound images. We created a dataset relating depth images to carcass traits obtained from trained-evaluators ultrasound measurements of two thousand animals from Brazilian live cattle. We then modeled CNNs able to estimate such carcass traits. Prior CNN results have produced mean square errors of ~10e-4. It suggests that such combined machine learning and computer vision approach can potentially replace ultrasound measurements.
11843-45
Author(s): Abu Taufique, Andreas Savakis, Rochester Institute of Technology (United States); Michael Braun, Daniel Kubacki, Ethan Dell, Systems & Technology Research (United States); Lei Qian, Sean O'Rourke, Air Force Research Lab. (United States)
Show Abstract + Hide Abstract
Siamese deep-network trackers have received significant attention in recent years due to their real-time speed and state-of-the-art performance. However, Siamese trackers suffer from similar looking confusers, that are prevalent in aerial imagery and create challenging conditions due to prolonged occlusions where the tracker object re-appears under different pose and illumination. Our work proposes SiamReID, a novel re-identification framework for Siamese trackers, that incorporates confuser rejection during prolonged occlusions and is well-suited for aerial tracking. The re-identification feature is trained using both triplet loss and a class balanced loss. Our approach achieves state-of-the-art performance in the UAVDT single object tracking benchmark.
11843-46
Author(s): Xiao Fu, Zhejiang Univ. (China); Tian-run Chen, Aoshen Labs Tech. (Hangzhou) Co., Ltd. (China); Juncheng Jiang, Zhejiang Univ. (China); Jia Zhang, Yangzhou Institute of Technology (China)
Show Abstract + Hide Abstract
In this research we explored acquiring 3D information of small objects based solely on the single 2D image of an object. We propose a single-view, single-shot 3D depth reconstruction method enabled by deep learning. A novel neural network was proposed that combines the low-level interpretation of image cues and high-level reasoning. While most existing methods focus on large scale 3D scene prediction or large objects 3D reconstruction, our neural network was tweaked to reach optimal results in reconstructing the depth of small objects, as we show in our preliminary results. An example of the proposed method applying to fabrication machines, such as CNC miller and laser engraver, to acquire detailed height map of pieces with nonmonotonous surface is demonstrated, by using an only build-in camera of the machine that is originally designed for monitoring the fabrication process.
11843-47
Author(s): Zhaowei Chen, Samuel J. Vidourek, Mikey R. Holtz, Hossein Alisafaee, Rose-Hulman Institute of Technology (United States)
Show Abstract + Hide Abstract
We have designed and built a machine vision system for the inspection and monitoring at the production line of fiber interlock armor cables. Our economical solution for the design approach has been to utilize a vision system based on a single camera and a tunnel source of uniform illumination in conjunction with two flat mirrors to obtain a 360-degree view of the product. The resolution of imaging system allows the detection of features on the order of tens of microns. The measurement and imperfection method utilizes a deep learning algorithm to detect manufacturing defects in the cable in-line with the production. Our vision system is able to inspect a variety of interlock armor cables with different sizes and shapes, making it uniquely versatile.
11843-48
Author(s): Viviane Oliveira Das Merces, Univ. Federal da Bahia (Brazil); Anderson Dourado Sisnando, Univ. Federal do Recôncavo of Bahia (Brazil); Vitaly Felix Rodriguez-Esquerre, Escola Politécnica da Univ. Federal da Bahia (Brazil)
Show Abstract + Hide Abstract
The present work deals with the implementation of machine learning algorithms for the analysis of the coupling efficiency of tapers for silicon photonics applications operating in the C band. The analyzed tapers are used for coupling a continuous waveguide with a periodical subwavelength waveguide and they are composed by several segments with variable lengths. The training, testing and validating data sets have been numerically obtained by an efficient frequency domain finite element method which solves the wave equation and determines the spatial distribution of the electromagnetic fields and the coupling efficiency for each taper configuration. An excellent agreement has been observed for the coupling efficiency calculation using the machine learning algorithms when compared with the one obtained by using the finite element method.
11843-49
Author(s): Zhicheng Cao, Xidian Univ. (China); Xing Chen, Siemens Ltd. (China); Shufen Cao, Case Western Reserve Univ. (United States); Liaojun Pang, Xidian Univ. (China)
11843-51
Author(s): Zhicheng Cao, Heng Zhao, Xidian Univ. (China); Shufen Cao, Case Western Reserve Univ. (United States); Liaojun Pang, Xidian Univ. (China)
Signal, Image, and Data Processing Plenary Session
Machine Vision and Manufacturing
Remote Sensing
11843-10
Author(s): Thomas Chen, The Academy for Math, Science and Engineering (United States)
Digital Forum: On-demand starting 1 August
Show Abstract + Hide Abstract
Natural disasters ravage the world's cities, valleys, and shores on a monthly basis. Using a dataset that includes labeled pre- and post- disaster satellite imagery, we train multiple convolutional neural networks to assess building damage after these devastating events on a per-building basis. In order to investigate how to best classify building damage, we present a highly interpretable deep-learning methodology that seeks to explicitly convey the most useful information required to train an accurate classification model. We also delve into which loss functions best optimize these models. Our findings include that ordinal-cross entropy loss is the most optimal loss function to use and that including the type of disaster that caused the damage in combination with a pre- and post-disaster image best predicts the level of damage caused. Our research seeks to computationally contribute to aiding in this ongoing and growing humanitarian crisis, heightened by climate change.
Algorithms and Imaging
11843-16
Author(s): Sria Biswas, Balasubramanyam Appina, Indian Institute of Information Technology, Design and Manufacturing, Kancheepuram (India); Roopak R. Tamboli, Indian Institute of Technology Hyderabad (India); Peter Andras Kara, Budapest Univ. of Technology and Economics (Hungary); Aniko Simon, Sigma Technology (Hungary)
Digital Forum: On-demand starting 1 August
Show Abstract + Hide Abstract
In this paper, we introduce an exhaustive analysis regarding the practical applications of objective quality metrics for stereoscopic 3D imaging. Our contribution addresses each and every state-of-the-art objective metric in the scientific literature, separately for image and video quality. The study differentiates the metrics by input requirements and supervision, and examines performance via statistical measures. The paper focuses on the actual practical applications of the predictive models, and highlights relevant criteria, along with general feasibility, suitability and usability. The analysis of the investigated use cases also addresses potential future research questions and specifies the appropriate directives for quality-focused, user-centric development.
11843-18
Author(s): James P. Theiler, Christopher X. Ren, Los Alamos National Lab. (United States)
Digital Forum: On-demand starting 1 August
Show Abstract + Hide Abstract
Gaussianization is a recently suggested approach for density estimation from data drawn from a decidedly non-Gaussian, and possibly high dimensional, distribution. The key idea is to learn a transformation that, when applied to the data, leads to an approximately Gaussian distribution. The density, for any given point in the original distribution, is then given by the determinant of the transformation's Jacobian at that point, multiplied by the (analytically known) density of the Gaussian for the transformed data. In this work, we investigate the use of distilled machine learning to provide a compact implementation of the Gaussianization transform (which in usual practice is obtained iteratively), thereby enabling faster computation, better controlled regularization, and more direct estimation of the Jacobian. While density estimation underlies many statistical analyses, our interest is in hyperspectral detection problems.
Big Data, Simulations, and Physics
Consumer Applications and E-Commerce
Poster Session
11843-36
Author(s): Neha Konakalla, Sai Charan Parasharam, Sridhar Varadala, Vidya Jyothi Institute of Technology (India)
Digital Forum: On-demand starting 1 August
11843-39
Author(s): Venkata Sai Krithik Pothula, Adarsh Raghavan Madabhushi Tirumala , Sai Pratyusha Kundeti, Vasanth K., Mahesh Rajendran, Anand Pandarinath Madanwad, Vidya Jyothi Institute of Technology (India)
Digital Forum: On-demand starting 1 August
Show Abstract + Hide Abstract
Research from WHO shows in every 1.44 minutes a teenager is killed in road traffic crashes around the world. Our paper suggests the use of Deep Learning Algorithms in preventing such casualties through hardware and software implementation of the device in motor vehicles, also to overcome the potential limitation of Face Recognition to go that extra mile. We suggest using the integration of 4 models namely, Face Detection, Passive Liveliness Detection(PLD), Face Recognition, Eye Detection. PLD model is used for Presentation Attack Detection. The face recognition model is trained using the shape predictor 68 landmarks which are unique for each face. Using these landmarks, monitoring of the sleepiness is carried out. Along with that different current attack scenarios/ limitations of Facial Recognition that will be faced with these devices are described. Based on these scenarios, some of the preventive methods are elaborated to make the purpose of the device to its fullest performance.
11843-50
Author(s): Vadim Dubovskov, Alexander A. Zelensky, Moscow State Univ. of Technology "Stankin" (Russian Federation)
Digital Forum: On-demand starting 1 August
Show Abstract + Hide Abstract
The article attempts to develop a control system for a manipulator of a parallel structure with flexible connections to create an industrial manipulator. The implementation of the manipulator control system involves the use of a digital differential analyzer algorithm, which is widely used in numerical control systems. The trajectory of movement of the working tool of the manipulator can be specified in the form of a text file of the program in G-codes or using the information and control shell of the manipulator control system. A feature of the control system under study is the presence of a machine vision unit used to set the position of the working tool of the manipulator in space.
Conference Chair
Lawrence Livermore National Lab. (United States)
Conference Chair
Univ. of Dayton (United States)
Conference Chair
NVIDIA Corp. (United States)
Conference Co-Chair
Lawrence Livermore National Lab. (United States)
Conference Co-Chair
Old Dominion Univ. (United States)
Program Committee
St. Jude Children's Research Hospital (United States)
Program Committee
BeamIO (United States)
Program Committee
NVIDIA Corp. (United States)
Program Committee
Pyxeda, Inc. (United States)
Program Committee
Lawrence Livermore National Lab. (United States)
Program Committee
Lawrence Livermore National Lab. (United States)
Program Committee
Univ. of Texas San Antonio Chapter (United States)
Program Committee
Lawrence Livermore National Lab. (United States)
Program Committee
Univ. of Dayton (United States)
Program Committee
Etsy, Inc. (United States)
Program Committee
Lawrence Livermore National Lab. (United States)
Program Committee
Manar D. Samad
Tennessee State Univ. (United States)
Program Committee
ArchiFiction, Inc. (United States)
Program Committee
Los Alamos National Lab. (United States)
Additional Information

POST-DEADLINE SUBMISSIONS

  • Submissions accepted through 15-June
  • Notification of acceptance by: 1-July

View Call for Papers PDF