Learning and then recognizing a route, whether travelled during the day or at night, in clear or inclement weather, and in summer or winter is a challenging task for state of the art algorithms in computer vision and robotics. In this paper, we present a new approach to visual navigation under changing conditions dubbed SeqSLAM. Instead of calculating the single location most likely given a current image, our approach calculates the best candidate matching location within every local navigation sequence. Localization is then achieved by recognizing coherent sequences of these “local best matches”. This approach removes the need for global matching performance by the vision front-end - instead it must only pick the best match within any short sequence of images. The approach is applicable over environment changes that render traditional feature-based techniques ineffective. Using two car-mounted camera datasets we demonstrate the effectiveness of the algorithm and compare it to one of the most successful feature-based SLAM algorithms, FAB-MAP. The perceptual change in the datasets is extreme; repeated traverses through environments during the day and then in the middle of the night, at times separated by months or years and in opposite seasons, and in clear weather and extremely heavy rain. While the feature-based method fails, the sequence-based algorithm is able to match trajectory segments at 100% precision with recall rates of up to 60%.

SeqSLAM : visual route-based navigation for sunny summer days and stormy winter nights

Proceedings of world academy of science, engineering and technology, vol 8

Compressed Sensing MRI (CS-MRI) has provided theoretical foundations upon which the time-consuming MRI acquisition process can be accelerated. However, it primarily relies on iterative numerical solvers which still hinders their adaptation in time-critical applications. In addition, recent advances in deep neural networks have shown their potential in computer vision and image processing, but their adaptation to MRI reconstruction is still in an early stage. In this paper, we propose a novel deep learning-based generative adversarial model, RefineGAN, for fast and accurate CS-MRI reconstruction. The proposed model is a variant of fully-residual convolutional autoencoder and generative adversarial networks (GANs), specifically designed for CS-MRI formulation; it employs deeper generator and discriminator networks with cyclic data consistency loss for faithful interpolation in the given under-sampled k-space data. In addition, our solution leverages a chained network to further enhance the reconstruction quality. RefineGAN is fast and accurate -- the reconstruction process is extremely rapid, as low as tens of milliseconds for reconstruction of a 256x256 image, because it is one-way deployment on a feed-forward network, and the image quality is superior even for extremely low sampling rate (as low as 10%) due to the data-driven nature of the method. We demonstrate that RefineGAN outperforms the state-of-the-art CS-MRI methods by a large margin in terms of both running time and image quality via evaluation using several open-source MRI databases.

Compressed Sensing MRI Reconstruction using a Generative Adversarial Network with a Cyclic Loss

MRI based medical image analysis: Survey on brain tumor grade classification

Deep learning approaches to biomedical image segmentation

Review of MRI-based Brain Tumor Image Segmentation Using Deep Learning Methods

This paper proposes a novel improvement of forecasting approach based on using time-invariant fuzzy time series. In contrast to traditional forecasting methods, fuzzy time series can be also applied to problems, in which historical data are linguistic values. It is shown that proposed time-invariant method improves the performance of forecasting process. Further, the effect of using different number of fuzzy sets is tested as well. As with the most of cited papers, historical enrollment of the University of Alabama is used in this study to illustrate the forecasting process. Subsequently, the performance of the proposed method is compared with existing fuzzy time series time-invariant models based on forecasting accuracy. It reveals a certain performance superiority of the proposed method over methods described in the literature.

/pdf/forecasting-enrollment-model-based-on-first-order-fuzzy-time-fd6em91kto.pdf

Forecasting Enrollment Model Based on First-Order Fuzzy Time Series

In this paper, we propose a novel optical flow based features for abnormal crowd behaviour detection. The proposed feature is mainly based on the angle difference computed between the optical flow vectors in the current frame and in the previous frame at each pixel location. The angle difference information is also combined with the optical flow magnitude to produce new, effective and direction invariant event features. A one-class SVM is utilized to learn normal crowd behavior. If a test sample deviates significantly from the normal behavior, it is detected as abnormal crowd behavior. Although there are many optical flow based features for crowd behaviour analysis, this is the first time the angle difference between optical flow vectors in the current frame and in the previous frame is considered as a anomaly feature. Evaluations on UMN and PETS2009 datasets show that the proposed method performs competitive results compared to the state-of-the-art methods.

/pdf/abnormal-crowd-behavior-detection-using-novel-optical-flow-1ga9jjed98.pdf

Abnormal crowd behavior detection using novel optical flow-based features

Enterprises provide professionally authored content about their products/services in different languages for use in web sites and customer care. For customer care, personalization/personalized information delivery is becoming important since it re-encourages users to return to the service provider. Personalization usually requires both contextual and descriptive metadata. But current metadata authored by content developers is usually quite simple. In this paper, we introduce an automatic metadata extraction framework, which can extract multilingual metadata from the enterprise content, for a personalized information retrieval system. We introduce two new ontologies for metadata creation and a novel semi-automatic topic vocabulary extraction algorithm. We demonstrate and evaluate our approach on the English and German Symantec Norton 360 technical content. Evaluations indicate that the proposed approach produces rich and high quality metadata for a personalized information retrieval system.

/pdf/automatic-metadata-extraction-from-multilingual-enterprise-351stp4baf.pdf

Automatic metadata extraction from multilingual enterprise content

Monitoring continuously captured surveillance videos is a challenging and time consuming task. To assist this issue, a new framework is introduced that applies anomaly detection, semantic annotation and provides a concept-based search interface. In particular, novel optical flow based features are used for abnormal crowd behaviour detection. Then, processed surveillance videos are annotated using a new semantic metadata model based on multimedia standards using Semantic Web technologies. In this way, globally inter-operable metadata about abnormal crowd behaviours are generated. Finally, for the first time, based on crowd behaviours, a novel concept-based semantic search interface is proposed. In the proposed interface, along with search results (video segments), statistical data about crowd behaviours are also presented. With extensive user studies, it is demonstrated that the proposed concept-based semantic search interface enables efficient search and analysis of abnormal crowd behaviours. Although there are existing works to achieve (a) crowd anomaly detection, (b) semantic annotation and (c) semantic search interface, none of the existing works combine these three system components in a novel framework like the one proposed in this paper. In each system component, we introduce contributions to the field as well as use the Semantic Web technologies to combine and standardize output of different system components; output of the anomaly detection is automatically annotated with metadata and stored to a semantic database. When continuous surveillance videos are processed, only the semantic database is updated. Finally, the user interface queries the updated database for searching/analyzing surveillance videos without changing any coding. Thus, the framework supports re-usability. This paper explains and evaluates different components of the framework.

Melike Sah

Papers

Review of MRI-based Brain Tumor Image Segmentation Using Deep Learning Methods

Forecasting Enrollment Model Based on First-Order Fuzzy Time Series

Abnormal crowd behavior detection using novel optical flow-based features

Automatic metadata extraction from multilingual enterprise content

A novel framework and concept-based semantic search Interface for abnormal crowd behaviour analysis in surveillance videos