Signal Processing

      The following dataset consists of utterances, recorded using 24 volunteers raised in the Province of Manitoba, Canada. To provide a repeatable set of test words that would cover all of the phonemes, the Edinburg Machine Readable Phonetic Alphabet (MRPA) [KiGr08], consisting of 44 words is used. Each recording consists of one word uttered by the volunteer and recorded in one continuous session.

365 views
  • Signal Processing
  • Last Updated On: 
    Thu, 11/08/2018 - 10:34

    This dataset is associated with the paper, Jackson & Hall 2016, which is open source, and can be found here: http://ieeexplore.ieee.org/document/7742994/

    The DataPort Repository contains the data used primarily for generating Figure 1.

    518 views
  • Biomedical and Health Sciences
  • Last Updated On: 
    Sat, 06/16/2018 - 23:05

    The distributed generation, along with the deregulation of the Smart Grid, have created a great concern on Power Quality (PQ), as it has a direct impact on utilities and customers, as well as effects on the sinusoidal signal of the power line. The a priori unknown features of the distributed energy resources (DER) introduce non-linear behaviours in loads associated to a variety of PQ disturbances.

    1395 views
  • Power and Energy
  • Last Updated On: 
    Thu, 11/08/2018 - 10:34

    Low light scenes often come with acquisition noise, which not only disturbs the viewers, but it also makes video compression harder. These type of videos are often encountered in cinema as a result of artistic perspective or the nature of a scene. Other examples include shots of wildlife (e.g. mobula rays at night in Blue Planet II), concerts and shows, surveillance camera footage and more. Inspired by all above, we are proposing a challenge on encoding low-light captured videos.

  • Image Processing
  • Signal Processing
  • Other
  • Last Updated On: 
    Sat, 11/09/2019 - 13:17

    The sensitivity of an ultrasonic transducer is an important parameter for evaluating its effective frequency band, electroacoustic conversion efficiency, and the measurement capability of the system. Determining the sensitivity of a traditional immersion or contact piezoelectric transducer has been well investigated. However, due to the high attenuation of wave propagation in air and the large acoustic impedance mismatch between the active piezoceramic material and the load medium, there are few reports on the calibration of an air-coupled piezoelectric transducer.

    21 views
  • Image Processing
  • Last Updated On: 
    Mon, 11/04/2019 - 03:38

     Data processing procedure. Pictures from (a) to (d) are corresponding to the four data processing steps respectively. The units of vertical axis are arbitrary and the number is not equal to the actual received power. The measurement range of 400 m corresponds to the beam angle of about 46°

    36 views
  • Signal Processing
  • Last Updated On: 
    Tue, 10/29/2019 - 04:14

    In order to increase the diversity in signal datasets, we create a new dataset called HisarMod, which includes 26 classes and 5 different modulation families passing through 5 different wireless communication channel. During the generation of the dataset, MATLAB 2017a is employed for creating random bit sequences, symbols, and wireless fading channels. 

     

    105 views
  • Artificial Intelligence
  • Last Updated On: 
    Tue, 11/12/2019 - 08:05

    Oxygen is one of the most adverse gas that contaminates sterile drugs in glass medicine bottle, so it is of great significance to detect oxygen concentration for glass medicine bottle in order to ensure the asepsis of drug and the stability of ingredients. Wavelength modulation spectroscopy (WMS) is applied to achieve online oxygen concentration detection by the single-line spectrum analysis for the advantages of non-contact and high sensitivity.

    34 views
  • Signal Processing
  • Last Updated On: 
    Sat, 10/19/2019 - 09:05

    7200 .csv files, each containing a 10 kHz recording of a 1 ms lasting 100 hz sound, recorded centimeterwise in a 20 cm x 60 cm locating range on a table. 3600 files (3 at each of the 1200 different positions) are without an obstacle between the loudspeaker and the microphone, 3600 RIR recordings are affected by the changes of the object (a book). The OOLA is initially trained offline in batch mode by the first instance of the RIR recordings without the book. Then it learns online in an incremental mode how the RIR changes by the book.

    123 views
  • Artificial Intelligence
  • Last Updated On: 
    Mon, 11/04/2019 - 07:37

    As one of the research directions at OLIVES Lab @ Georgia Tech, we focus on the robustness of data-driven algorithms under diverse challenging conditions where trained models can possibly be depolyed. To achieve this goal, we introduced a large-sacle (1.M images) object recognition dataset (CURE-OR) which is among the most comprehensive datasets with controlled synthetic challenging conditions. In CURE

    132 views
  • Artificial Intelligence
  • Last Updated On: 
    Sun, 10/13/2019 - 17:05

    Pages