AI, Volume 1, Issue 4 (December 2020) – 7 articles

To judge the ability of convolutional neural networks (CNNs) to effectively and efficiently transfer image representations learned on the ImageNet dataset to the task of recognizing COVID-19 in this work, we propose and analyze four approaches. For this purpose, we use VGG16, ResNetV2, InceptionResNetV2, DenseNet121, and MobileNetV2 CNN models pre-trained on ImageNet dataset to extract features from X-ray images of COVID and Non-COVID patients. Simulations study performed by us reveal that these pre-trained models have a different level of ability to transfer image representation. We find that in the approaches that we have proposed, if we use either ResNetV2 or DenseNet121 to extract features, then the performance of these approaches to detect COVID-19 is better. One of the important findings of our study is that the use of principal component analysis for feature selection improves efficiency. The approach using the fusion of features outperforms all the other approaches, and with this approach, we could achieve an accuracy of 0.94 for a three-class classification problem. This work will not only be useful for COVID-19 detection but also for any domain with small datasets. Full article

This is motivated by a requirement for robust, autonomy-enabling scene understanding in unknown environments. In the method proposed in this paper, discriminative machine-learning approaches are applied to infer traversability and predict routes from Frequency-Modulated Contunuous-Wave (FMCV) radar frames. Firstly, using geometric features extracted from LiDAR point clouds as inputs to a fuzzy-logic rule set, traversability pseudo-labels are assigned to radar frames from which weak supervision is applied to learn traversability from radar. Secondly, routes through the scanned environment can be predicted after they are learned from the odometry traces arising from traversals demonstrated by the autonomous vehicle (AV). In conjunction, therefore, a model pretrained for traversability prediction is used to enhance the performance of the route proposal architecture. Experiments are conducted on the most extensive radar-focused urban autonomy dataset available to the community. Our key finding is that joint learning of traversability and demonstrated routes lends itself best to a model which understands where the vehicle should feasibly drive. We show that the traversability characteristics can be recovered satisfactorily, so that this recovered representation can be used in optimal path planning, and that an end-to-end formulation including both traversability feature extraction and routes learned by expert demonstration recovers smooth, drivable paths that are comprehensive in their coverage of the underlying road network. We conclude that the proposed system will find use in enabling mapless vehicle autonomy in extreme environments. Full article

The coronavirus disease 2019 (COVID-19) global pandemic has severely impacted lives across the globe. Respiratory disorders in COVID-19 patients are caused by lung opacities similar to viral pneumonia. A Computer-Aided Detection (CAD) system for the detection of COVID-19 using chest radiographs would provide a second opinion for radiologists. For this research, we utilize publicly available datasets that have been marked by radiologists into two-classes (COVID-19 and non-COVID-19). We address the class imbalance problem associated with the training dataset by proposing a novel transfer-to-transfer learning approach, where we break a highly imbalanced training dataset into a group of balanced mini-sets and apply transfer learning between these. We demonstrate the efficacy of the method using well-established deep convolutional neural networks. Our proposed training mechanism is more robust to limited training data and class imbalance. We study the performance of our algorithm(s) based on 10-fold cross validation and two hold-out validation experiments to demonstrate its efficacy. We achieved an overall sensitivity of 0.94 for the hold-out validation experiments containing 2265 and 2139 marked as COVID-19 chest radiographs, respectively. For the 10-fold cross validation experiment, we achieve an overall Area under the Receiver Operating Characteristic curve (AUC) value of 0.996 for COVID-19 detection. This paper serves as a proof-of-concept that an automated detection approach can be developed with a limited set of COVID-19 images, and in areas with scarcity of trained radiologists. Full article

We present an extensive evaluation of a wide variety of promising design patterns for automated deep-learning (AutoDL) methods, organized according to the problem categories of the 2019 AutoDL challenges, which set the task of optimizing both model accuracy and search efficiency under tight time and computing constraints. We propose structured empirical evaluations as the most promising avenue to obtain design principles for deep-learning systems due to the absence of strong theoretical support. From these evaluations, we distill relevant patterns which give rise to neural network design recommendations. In particular, we establish (a) that very wide fully connected layers learn meaningful features faster; we illustrate (b) how the lack of pretraining in audio processing can be compensated by architecture search; we show (c) that in text processing deep-learning-based methods only pull ahead of traditional methods for short text lengths with less than a thousand characters under tight resource limitations; and lastly we present (d) evidence that in very data- and computing-constrained settings, hyperparameter tuning of more traditional machine-learning methods outperforms deep-learning systems. Full article

The natural environment and our interaction with it are essentially multisensory, where we may deploy visual, tactile and/or auditory senses to perceive, learn and interact with our environment. Our objective in this study is to develop a scene analysis algorithm using multisensory information, specifically vision and audio. We develop a proto-object-based audiovisual saliency map (AVSM) for the analysis of dynamic natural scenes. A specialized audiovisual camera with 360${}^{\circ }$ field of view, capable of locating sound direction, is used to collect spatiotemporally aligned audiovisual data. We demonstrate that the performance of a proto-object-based audiovisual saliency map in detecting and localizing salient objects/events is in agreement with human judgment. In addition, the proto-object-based AVSM that we compute as a linear combination of visual and auditory feature conspicuity maps captures a higher number of valid salient events compared to unisensory saliency maps. Such an algorithm can be useful in surveillance, robotic navigation, video compression and related applications. Full article

Digital images often become corrupted by undesirable noise during the process of acquisition, compression, storage, and transmission. Although the kinds of digital noise are varied, current denoising studies focus on denoising only a single and specific kind of noise using a devoted deep-learning model. Lack of generalization is a major limitation of these models. They cannot be extended to filter image noises other than those for which they are designed. This study deals with the design and training of a generalized deep learning denoising model that can remove five different kinds of noise from any digital image: Gaussian noise, salt-and-pepper noise, clipped whites, clipped blacks, and camera shake. The denoising model is constructed on the standard segmentation U-Net architecture and has three variants—U-Net with Group Normalization, Residual U-Net, and Dense U-Net. The combination of adversarial and L1 norm loss function re-produces sharply denoised images and show performance improvement over the standard U-Net, Denoising Convolutional Neural Network (DnCNN), and Wide Interface Network (WIN5RB) denoising models. Full article

Figure Ground Organization (FGO)-inferring spatial depth ordering of objects in a visual scene-involves determining which side of an occlusion boundary is figure (closer to the observer) and which is ground (further away from the observer). A combination of global cues, like convexity, and local cues, like T-junctions are involved in this process. A biologically motivated, feed forward computational model of FGO incorporating convexity, surroundedness, parallelism as global cues and spectral anisotropy (SA), T-junctions as local cues is presented. While SA is computed in a biologically plausible manner, the inclusion of T-Junctions is biologically motivated. The model consists of three independent feature channels, Color, Intensity and Orientation, but SA and T-Junctions are introduced only in the Orientation channel as these properties are specific to that feature of objects. The effect of adding each local cue independently and both of them simultaneously to the model with no local cues is studied. Model performance is evaluated based on figure-ground classification accuracy (FGCA) at every border location using the BSDS 300 figure-ground dataset. Each local cue, when added alone, gives statistically significant improvement in the FGCA of the model suggesting its usefulness as an independent FGO cue. The model with both local cues achieves higher FGCA than the models with individual cues, indicating SA and T-Junctions are not mutually contradictory. Compared to the model with no local cues, the feed-forward model with both local cues achieves $\ge 8.78$% improvement in terms of FGCA. Full article