KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Subject: search.filters.subject.deep neural networks

Search Results

Now showing 1 - 3 of 3
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Edge-Driven Multi-Agent Reinforcement Learning: A Novel Approach to Ultrasound Breast Tumor Segmentation
    (2023-12-01)
    Karunanayake, Nalan
    ;
    Moodleah, Samart
    ;
    Makhanov, Stanislav S.
    A segmentation model of the ultrasound (US) images of breast tumors based on virtual agents trained using reinforcement learning (RL) is proposed. The agents, living in the edge map, are able to avoid false boundaries, connect broken parts, and finally, accurately delineate the contour of the tumor. The agents move similarly to robots navigating in the unknown environment with the goal of maximizing the rewards. The individual agent does not know the goal of the entire population. However, since the robots communicate, the model is able to understand the global information and fit the irregular boundaries of complicated objects. Combining the RL with a neural network makes it possible to automatically learn and select the local features. In particular, the agents handle the edge leaks and artifacts typical for the US images. The proposed model outperforms 13 state-of-the-art algorithms, including selected deep learning models and their modifications.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    A DNN-Based Accurate Masking Using Significant Feature Sets
    (2022-01-01)
    Sivapatham, Shoba
    ;
    Goel, Pankaj
    ;
    Burra, Srikanth
    ;
    Sooraksa, Pitikhate
    ;
    Kar, Asutosh
    Monaural speech separation has remained a very challenging problem for a longtime which can be addressed using a supervised learning approach that uses features of the noisy input to predict an accurate time-frequency mask. Effective acoustic phonetic features can help in the accurate mask prediction at low Signal-to-Noise Ratios (SNRs). Individual features capture specific attributes of the audio signal; therefore, it's essential to employ a set of features. This work examines different combinations of monaural features as input and ideal ratio mask a straining target to the DNN model. Feature combination sets are constructed by examining single features and then combining the most relevant ones. The results are evaluated for different feature combinations under non-stationary noises at low SNR levels. The feature performance is evaluated by using intelligibility and quality measures. A combination of two features is considered the best feature combination as it indicates a significant increase in speech intelligibility as compared to individual features and combinations consisting of more than two features.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Deep Neural Networks for Sound Synthesis of Thai Duct F1ute, Khlui
    (2021-04-01)
    Sinjankhom, Tantep
    ;
    Chivapreecha, Sorawat
    ;
    Chitanont, Nachanant
    ;
    Kato, Tomonori
    This paper introduces Thai musical instrument sound synthesis with an intelligent technique which is a combination of deep neural networks and conventional signal processing techniques. The instrument focused on in this paper is the Khlui, which is a unique Thai reedless wind instrument. Khlui sound can be synthesized by performing a combination of additive and subtractive syntheses. The synthesis system takes the pitch and loudness of any monophonic input signal. Then, multilayer perceptron and recurrent neural networks are used together in order to generate 3 parameters which are harmonic distribution, filter magnitude response, and the signal envelope. Finally, the harmonic characteristic of the Khlui along with the blowing noises are highly synthesized. The results are natural-sounding and realistic when compared with the recorded Khlui audio. This research will contribute to the more convenient natural synthesis of the Khlui sounds.