Search results for: audio segmentation

Authors: Xu Han, Shanxiong Chen, Shiyu Zhu, Xiaoyu Lin, Fujia Zhao, Dingwang Wang

Abstract:

Character detection is an important issue for character recognition of ancient Yi books. The accuracy of detection directly affects the recognition effect of ancient Yi books. Considering the complex layout, the lack of standard typesetting and the mixed arrangement between images and texts, we propose a character detection method for ancient Yi books based on connected components and regressive character segmentation. First, the scanned images of ancient Yi books are preprocessed with nonlocal mean filtering, and then a modified local adaptive threshold binarization algorithm is used to obtain the binary images to segment the foreground and background for the images. Second, the non-text areas are removed by the method based on connected components. Finally, the single character in the ancient Yi books is segmented by our method. The experimental results show that the method can effectively separate the text areas and non-text areas for ancient Yi books and achieve higher accuracy and recall rate in the experiment of character detection, and effectively solve the problem of character detection and segmentation in character recognition of ancient books.

Keywords: CCS concepts, computing methodologies, interest point, salient region detections, image segmentation

Procedia PDF Downloads 99

Authors: Larry Powell, Seth Polsley, Drew Casey, Tracy Hammond

Abstract:

Human activity recognition (HAR) systems have shown positive performance when recognizing repetitive activities like walking, running, and sleeping. Water-based activities are a reasonably new area for activity recognition. However, water-based activity recognition has largely focused on supporting the elite and competitive swimming population, which already has amazing coordination and proper form. Beginner swimmers are not perfect, and activity recognition needs to support the individual motions to help beginners. Activity recognition algorithms are traditionally built around short segments of timed sensor data. Using a time window input can cause performance issues in the machine learning model. The window’s size can be too small or large, requiring careful tuning and precise data segmentation. In this work, we present a method that uses a time window as the initial segmentation, then separates the data based on the change in the sensor value. Our system uses a multi-phase segmentation method that pulls all peaks and valleys for each axis of an accelerometer placed on the swimmer’s lower back. This results in high recognition performance using leave-one-subject-out validation on our study with 20 beginner swimmers, with our model optimized from our final dataset resulting in an F-Score of 0.95.

Keywords: time window, peak/valley segmentation, feature extraction, beginner swimming, activity recognition

Procedia PDF Downloads 85

Authors: Lae-Jeong Park

Abstract:

The paper presents a method that utilizes figure-ground color segmentation to extract effective global feature in terms of false positive reduction in the head-shoulder detection. Conventional detectors that rely on local features such as HOG due to real-time operation suffer from false positives. Color cue in an input image provides salient information on a global characteristic which is necessary to alleviate the false positives of the local feature based detectors. An effective approach that uses figure-ground color segmentation has been presented in an effort to reduce the false positives in object detection. In this paper, an extended version of the approach is presented that adopts separate multipart foregrounds instead of a single prior foreground and performs the figure-ground color segmentation with each of the foregrounds. The multipart foregrounds include the parts of the head-shoulder shape and additional auxiliary foregrounds being optimized by a search algorithm. A classifier is constructed with the feature that consists of a set of the multiple resulting segmentations. Experimental results show that the presented method can discriminate more false positive than the single prior shape-based classifier as well as detectors with the local features. The improvement is possible because the presented approach can reduce the false positives that have the same colors in the head and shoulder foregrounds.

Keywords: pedestrian detection, color segmentation, false positive, feature extraction

Procedia PDF Downloads 255

Authors: Abhisek Dash

Abstract:

This study had three neural networks, one for number segmentation, one for number detection and one for number recognition all of which are coupled to one another. All networks were trained on the MNIST dataset and were convolutional. It was assumed that the images had lighter background and darker foreground. The segmentation network took 28x28 images as input and had sixteen outputs. Segmentation training starts when a dark pixel is encountered. Taking a window(7x7) over that pixel as focus, the eight neighborhood of the focus was checked for further dark pixels. The segmentation network was then trained to move in those directions which had dark pixels. To this end the segmentation network had 16 outputs. They were arranged as “go east”, ”don’t go east ”, “go south east”, “don’t go south east”, “go south”, “don’t go south” and so on w.r.t focus window. The focus window was resized into a 28x28 image and the network was trained to consider those neighborhoods which had dark pixels. The neighborhoods which had dark pixels were pushed into a queue in a particular order. The neighborhoods were then popped one at a time stitched to the existing partial image of the number one at a time and trained on which neighborhoods to consider when the new partial image was presented. The above process was repeated until the image was fully covered by the 7x7 neighborhoods and there were no more uncovered black pixels. During testing the network scans and looks for the first dark pixel. From here on the network predicts which neighborhoods to consider and segments the image. After this step the group of neighborhoods are passed into the detection network. The detection network took 28x28 images as input and had two outputs denoting whether a number was detected or not. Since the ground truth of the bounds of a number was known during training the detection network outputted in favor of number not found until the bounds were not met and vice versa. The recognition network was a standard CNN that also took 28x28 images and had 10 outputs for recognition of numbers from 0 to 9. This network was activated only when the detection network votes in favor of number detected. The above methodology could segment connected and overlapping numbers. Additionally the recognition unit was only invoked when a number was detected which minimized false positives. It also eliminated the need for rules of thumb as segmentation is learned. The strategy can also be extended to other characters as well.

Keywords: convolutional neural networks, OCR, text detection, text segmentation

Procedia PDF Downloads 130

Authors: Sadia Basar, Naila Habib, Awais Adnan

Abstract:

The paper presents the Technique for Image Segmentation Using K-Means Clustering Classification. The presented algorithms were specific, however, missed the neighboring information and required high-speed computerized machines to run the segmentation algorithms. Clustering is the process of partitioning a group of data points into a small number of clusters. The proposed method is content-aware and feature extraction method which is able to run on low-end computerized machines, simple algorithm, required low-quality streaming, efficient and used for security purpose. It has the capability to highlight the boundary and the object. At first, the user enters the data in the representation of the input. Then in the next step, the digital image is converted into groups clusters. Clusters are divided into many regions. The same categories with same features of clusters are assembled within a group and different clusters are placed in other groups. Finally, the clusters are combined with respect to similar features and then represented in the form of segments. The clustered image depicts the clear representation of the digital image in order to highlight the regions and boundaries of the image. At last, the final image is presented in the form of segments. All colors of the image are separated in clusters.

Keywords: clustering, image segmentation, K-means function, local and global minimum, region

Procedia PDF Downloads 350

Authors: Swathi Gopakumar, Sruthi Krishna, Shivasubramani Krishnamoorthy

Abstract:

Contactless, painless and radiation-free thermal imaging technology is one of the preferred screening modalities for detection of breast cancer. However, poor signal to noise ratio and the inexorable need to preserve edges defining cancer cells and normal cells, make the segmentation process difficult and hence unsuitable for computer-aided diagnosis of breast cancer. This paper presents key findings from a research conducted on the appraisal of two promising techniques, for the detection of breast cancer: (I) marker-controlled, Level-set segmentation of anisotropic diffusion filtered preprocessed image versus (II) Segmentation using marker-controlled level-set on a Gaussian-filtered image. Gaussian-filtering processes the image uniformly, whereas anisotropic filtering processes only in specific areas of a thermographic image. The pre-processed (Gaussian-filtered and anisotropic-filtered) images of breast samples were then applied for segmentation. The segmentation of breast starts with initial level-set function. In this study, marker refers to the position of the image to which initial level-set function is applied. The markers are generally placed on the left and right side of the breast, which may vary with the breast size. The proposed method was carried out on images from an online database with samples collected from women of varying breast characteristics. It was observed that the breast was able to be segmented out from the background by adjustment of the markers. From the results, it was observed that as a pre-processing technique, anisotropic filtering with level-set segmentation, preserved the edges more effectively than Gaussian filtering. Segmented image, by application of anisotropic filtering was found to be more suitable for feature extraction, enabling automated computer-aided diagnosis of breast cancer.

Keywords: anisotropic diffusion, breast, Gaussian, level-set, thermograms

Procedia PDF Downloads 353

Authors: Abder-Rahman Ali, Adélaïde Albouy-Kissi, Manuel Grand-Brochier, Viviane Ladan-Marcus, Christine Hoeffl, Claude Marcus, Antoine Vacavant, Jean-Yves Boire

Abstract:

In this paper, we present a new segmentation approach for focal liver lesions in contrast enhanced ultrasound imaging. This approach, based on a two-cluster Fuzzy C-Means methodology, considers type-II fuzzy sets to handle uncertainty due to the image modality (presence of speckle noise, low contrast, etc.), and to calculate the optimum inter-cluster threshold. Fine boundaries are detected by a local recursive merging of ambiguous pixels. The method has been tested on a representative database. Compared to both Otsu and type-I Fuzzy C-Means techniques, the proposed method significantly reduces the segmentation errors.

Keywords: defuzzification, fuzzy clustering, image segmentation, type-II fuzzy sets

Procedia PDF Downloads 454

Authors: Leah Lee

Abstract:

This paper examines the correlation between deep learning models and emotions with noises to see whether or not noises mask emotions. The deep learning models used are plain convolutional neural networks (CNN), auto-encoder, long short-term memory (LSTM), and Visual Geometry Group-16 (VGG-16). Emotion datasets used are Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS), Crowd-sourced Emotional Multimodal Actors Dataset (CREMA-D), Toronto Emotional Speech Set (TESS), and Surrey Audio-Visual Expressed Emotion (SAVEE). To make it four times bigger, audio set files, stretch, and pitch augmentations are utilized. From the augmented datasets, five different features are extracted for inputs of the models. There are eight different emotions to be classified. Noise variations are white noise, dog barking, and cough sounds. The variation in the signal-to-noise ratio (SNR) is 0, 20, and 40. In summation, per a deep learning model, nine different sets with noise and SNR variations and just augmented audio files without any noises will be used in the experiment. To compare the results of the deep learning models, the accuracy and receiver operating characteristic (ROC) are checked.

Keywords: auto-encoder, convolutional neural networks, long short-term memory, speech emotion recognition, visual geometry group-16

Procedia PDF Downloads 38

Authors: Muhammad Zubair Khan, Yugyung Lee

Abstract:

Deep learning has recently achieved enormous response in semantic image segmentation. The previously developed U-Net inspired architectures operate with continuous stride and pooling operations, leading to spatial data loss. Also, the methods lack establishing long-term pixels connection to preserve context knowledge and reduce spatial loss in prediction. This article developed encoder-decoder architecture with bi-directional LSTM embedded in long skip-connections and densely connected convolution blocks. The network non-linearly combines the feature maps across encoder-decoder paths for finding dependency and correlation between image pixels. Additionally, the densely connected convolutional blocks are kept in the final encoding layer to reuse features and prevent redundant data sharing. The method applied batch-normalization for reducing internal covariate shift in data distributions. The empirical evidence shows a promising response to our method compared with other semantic segmentation techniques.

Keywords: deep learning, semantic segmentation, image analysis, pixels connection, convolution neural network

Procedia PDF Downloads 76

Authors: Ali Berkan Ural

Abstract:

This paper presents the analysis, evaluation, and pre-diagnosis of early stage breast based diagnostic problems (breast cancer, nodulesorlumps) by Computer Aided Diagnosing (CAD) system from mammogram radiological images. According to the statistics, the time factor is crucial to discover the disease in the patient (especially in women) as possible as early and fast. In the study, a new algorithm is developed using advanced image processing and deep learning method to detect and classify the problem at earlystagewithmoreaccuracy. This system first works with image processing methods (Image acquisition, Noiseremoval, Region Growing Segmentation, Morphological Operations, Breast BorderExtraction, Advanced Segmentation, ObtainingRegion Of Interests (ROIs), etc.) and segments the area of interest of the breast and then analyzes these partly obtained area for cancer detection/lumps in order to diagnosis the disease. After segmentation, with using the Spectrogramimages, 5 different deep learning based methods (specified Convolutional Neural Network (CNN) basedAlexNet, ResNet50, VGG16, DenseNet, Xception) are applied to classify the breast based problems.

Keywords: computer aided diagnosis, breast cancer, region growing, segmentation, deep learning

Procedia PDF Downloads 63

Authors: Marcio S. da Silva, Jose Mauricio de B. Bezerra, Antonio E. de A. Nogueira

Abstract:

This paper aims to make a performance analysis of shield wire transmission lines against atmospheric discharges when it is made the option of sectioning the shield wire and verify if the tolerability of the change. As a goal of this work, it was established to make complete modeling of a transmission line in the ATPDraw program with shield wire grounded in all the towers and in some towers. The methodology used to make the proposed evaluation was to choose an actual transmission line that served as a case study. From the choice of transmission line and verification of all its topology and materials, complete modeling of the line using the ATPDraw software was performed. Then several atmospheric discharges were simulated by striking the grounded shield wires in each tower. These simulations served to identify the behavior of the existing line against atmospheric discharges. After this first analysis, the same line was reconsidered with shield wire segmentation. The shielding wire segmentation technique aims to reduce induced losses in shield wires and is adopted in some transmission lines in Brazil. With the same conditions of atmospheric discharge the transmission line, this time with shield wire segmentation was again evaluated. The results obtained showed that it is possible to obtain similar performances against atmospheric discharges between a shield wired line in multiple towers and the same line with shield wire segmentation if some precautions are adopted as verification of the ground resistance of the wire segmented shield, adequacy of the maximum length of the segmented gap, evaluation of the separation length of the electrodes of the insulator spark, among others. As a conclusion, it is verified that since the correct assessment and adopted the correct criteria of adjustment a transmission line with shielded wire segmentation can perform very similar to the traditional use with multiple earths. This solution contributes in a very important way to the reduction of energy losses in transmission lines.

Keywords: atmospheric discharges, ATPDraw, shield wire, transmission lines

Procedia PDF Downloads 143

Authors: Mohamed A. Almoghalis, Osman M. Hegazy, Ibrahim F. Imam, Ali H. Elbastawessy

Abstract:

This paper presents a new hyperspectral image segmentation scheme with respect to both spatial and spectral contexts. The scheme uses the 8-pixels spatial pattern to build a weight structure that holds the number of outlier bands for each pixel among its neighborhood windows in different directions. The number of outlier bands for a pixel is obtained using bands quartile box plots profile among spatial 8-pixels pattern windows. The quartile box plot weight structure represents the spatial-spectral context in the image. Instead of starting segmentation process by single pixels, the proposed methodology starts by pixels groups that proved to share the same spectral features with respect to their spatial context. As a result, the segmentation scheme starts with Jigsaw pieces that build a mosaic image. The following step builds a model for each Jigsaw piece in the mosaic image. Each Jigsaw piece will be merged with another Jigsaw piece using KNN applied to their bands' quartile box plots profiles. The scheme iterates till required number of segments reached. Experiments use two data sets obtained from Earth Observer 1 (EO-1) sensor for Egypt and France. Initial results qualitative analysis showed encouraging results compared with ground truth. Quantitative analysis for the results will be included in the final paper.

Keywords: hyperspectral image segmentation, image processing, remote sensing, box plot

Procedia PDF Downloads 569

Authors: Mohammad Al-Dawoody

Abstract:

This study aims at exploring if priming is affected by gender in Modern Standard Arabic and if it is restricted solely to subjects with no specific language impairment (SLI). The sample in this study consists of 74 subjects, between the ages of 11;1 and 11;10, distributed into (a) 2 SLI experimental groups of 38 subjects divided into two gender groups of 18 females and 20 males and (b) 2 non-SLI control groups of 36 subjects divided into two gender groups of 17 females and 19 males. Employing a mixed research design, the researcher conducted this study within the framework of the relevance theory (RT) whose main assumption is that human beings are endowed with a biological ability to magnify the relevance of the incoming stimuli. Each of the four groups was given two different priming stimuli: audio-visual priming (T1) and syntactic priming (T2). The results showed that the priming effect was sheer distinct among SLI participants especially when retrieving typical responses (TR) in T1 and T2 with slight superiority of males over females. The results also revealed that non-SLI females showed stronger original response (OR) priming in T1 than males and that non-SLI males in T2 excelled in OR priming than females. Furthermore, the results suggested that the audio-visual priming has a stronger effect on SLI females than non-SLI females and that syntactic priming seems to have the same effect on the two groups (non-SLI and SLI females). The conclusion is that the priming effect varies according to gender and is not confined merely to non-SLI subjects.

Keywords: specific language impairment, relevance theory, audio-visual priming, syntactic priming, modern standard Arabic

Procedia PDF Downloads 143

Authors: Shafiullah Soomro

Abstract:

Active contour is one of the image segmentation techniques and its goal is to capture required object boundaries within an image. In this paper, we propose a novel image segmentation method by using an active contour method based on anisotropic diffusion feature enhancement technique. The traditional active contour methods use only pixel information to perform segmentation, which produces inaccurate results when an image has some noise or complex background. We use Perona and Malik diffusion scheme for feature enhancement, which sharpens the object boundaries and blurs the background variations. Our main contribution is the formulation of a new SPF (signed pressure force) function, which uses global intensity information across the regions. By minimizing an energy function using partial differential framework the proposed method captures semantically meaningful boundaries instead of catching uninterested regions. Finally, we use a Gaussian kernel which eliminates the problem of reinitialization in level set function. We use several synthetic and real images from different modalities to validate the performance of the proposed method. In the experimental section, we have found the proposed method performance is better qualitatively and quantitatively and yield results with higher accuracy compared to other state-of-the-art methods.

Keywords: active contours, anisotropic diffusion, level-set, partial differential equations

Procedia PDF Downloads 141

Authors: B. Kishore Kumar, Rakesh Pogula, T. Kishore Kumar

Abstract:

The steepness of an audio signal which is produced by the musical instruments, specifically percussive instruments is the perception of how high tone or low tone which can be considered as a frequency closely related to the fundamental frequency. This paper presents a novel method for silence removal and segmentation of music signals produced by the percussive instruments and the performance of proposed method is studied with the help of MATLAB simulations. This method is based on two simple features, namely the signal energy and the spectral centroid. As long as the feature sequences are extracted, a simple thresholding criterion is applied in order to remove the silence areas in the sound signal. The simulations were carried on various instruments like drum, flute and guitar and results of the proposed method were analyzed.

Keywords: percussive instruments, spectral energy, spectral centroid, silence removal

Procedia PDF Downloads 372

Authors: Harold Mauricio Díaz-Vargas, Cristian Alfonso Jimenez-Castaño, David Augusto Cárdenas-Peña, Guillermo Alberto Ortiz-Gómez, Alvaro Angel Orozco-Gutierrez

Abstract:

Precise identification of the nerves is a crucial task performed by anesthesiologists for an effective Peripheral Nerve Blocking (PNB). Now, anesthesiologists use ultrasound imaging equipment to guide the PNB and detect nervous structures. However, visual identification of the nerves from ultrasound images is difficult, even for trained specialists, due to artifacts and low contrast. The recent advances in deep learning make neural networks a potential tool for accurate nerve segmentation systems, so addressing the above issues from raw data. The most widely spread U-Net network yields pixel-by-pixel segmentation by encoding the input image and decoding the attained feature vector into a semantic image. This work proposes a conditioning approach and encoder pre-training to enhance the nerve segmentation of traditional U-Nets. Conditioning is achieved by the one-hot encoding of the kind of target nerve a the network input, while the pre-training considers five well-known deep networks for image classification. The proposed approach is tested in a collection of 619 US images, where the best C-UNet architecture yields an 81% Dice coefficient, outperforming the 74% of the best traditional U-Net. Results prove that pre-trained models with the conditional approach outperform their equivalent baseline by supporting learning new features and enriching the discriminant capability of the tested networks.

Keywords: nerve segmentation, U-Net, deep learning, ultrasound imaging, peripheral nerve blocking

Procedia PDF Downloads 76

Authors: Ali Hamza

Abstract:

Accurate segmentation of breast ultrasound images is of paramount importance in enhancing the diagnostic capabilities of breast cancer detection. This study presents an approach utilizing the U-Net architecture for segmenting breast ultrasound images aimed at improving the accuracy and reliability of mass identification within the breast tissue. The proposed method encompasses a multi-stage process. Initially, preprocessing techniques are employed to refine image quality and diminish noise interference. Subsequently, the U-Net architecture, a deep learning convolutional neural network (CNN), is employed for pixel-wise segmentation of regions of interest corresponding to potential breast masses. The U-Net's distinctive architecture, characterized by a contracting and expansive pathway, enables accurate boundary delineation and detailed feature extraction. To evaluate the effectiveness of the proposed approach, an extensive dataset of breast ultrasound images is employed, encompassing diverse cases. Quantitative performance metrics such as the Dice coefficient, Jaccard index, sensitivity, specificity, and Hausdorff distance are employed to comprehensively assess the segmentation accuracy. Comparative analyses against traditional segmentation methods showcase the superiority of the U-Net architecture in capturing intricate details and accurately segmenting breast masses. The outcomes of this study emphasize the potential of the U-Net-based segmentation approach in bolstering breast ultrasound image analysis. The method's ability to reliably pinpoint mass boundaries holds promise for aiding radiologists in precise diagnosis and treatment planning. However, further validation and integration within clinical workflows are necessary to ascertain their practical clinical utility and facilitate seamless adoption by healthcare professionals. In conclusion, leveraging the U-Net architecture for breast ultrasound image segmentation showcases a robust framework that can significantly enhance diagnostic accuracy and advance the field of breast cancer detection. This approach represents a pivotal step towards empowering medical professionals with a more potent tool for early and accurate breast cancer diagnosis.

Keywords: mage segmentation, U-Net, deep learning, breast cancer detection, diagnostic accuracy, mass identification, convolutional neural network

Procedia PDF Downloads 47

Authors: R. R. Ramsheeja, R. Sreeraj

Abstract:

For view the internal structures of the human body such as liver, brain, kidney etc have a wide range of different modalities for medical images are provided nowadays. Computer Tomography is one of the most significant medical image modalities. In this paper use CT liver images for study the use of automatic computer aided techniques to calculate the volume of the liver tumor. Segmentation method is used for the detection of tumor from the CT scan is proposed. Gaussian filter is used for denoising the liver image and Adaptive Thresholding algorithm is used for segmentation. Multiple Region Of Interest(ROI) based method that may help to characteristic the feature different. It provides a significant impact on classification performance. Due to the characteristic of liver tumor lesion, inherent difficulties appear selective. For a better performance, a novel proposed system is introduced. Multiple ROI based feature selection and classification are performed. In order to obtain of relevant features for Support Vector Machine(SVM) classifier is important for better generalization performance. The proposed system helps to improve the better classification performance, reason in which we can see a significant reduction of features is used. The diagnosis of liver cancer from the computer tomography images is very difficult in nature. Early detection of liver tumor is very helpful to save the human life.

Keywords: computed tomography (CT), multiple region of interest(ROI), feature values, segmentation, SVM classification

Procedia PDF Downloads 485

Authors: Pavel Chmelar, Martin Dobrovolny

Abstract:

Laser projection or laser footprint detection is today widely used in many fields of robotics, measurement, or electronics. The system accuracy strictly depends on precise laser footprint detection on target objects. This article deals with the laser line detection based on the RGB segmentation and the component labeling. As a measurement device was used the developed optical rangefinder. The optical rangefinder is equipped with vertical sweeping of the laser beam and high quality camera. This system was developed mainly for automatic exploration and mapping of unknown spaces. In the first section is presented a new detection algorithm. In the second section are presented measurements results. The measurements were performed in variable light conditions in interiors. The last part of the article present achieved results and their differences between day and night measurements.

Keywords: color segmentation, component labelling, laser line detection, automatic mapping, distance measurement, vector map

Procedia PDF Downloads 396

Authors: Donghee Noh, Seonhyeong Kim, Junhwan Choi, Heegon Kim, Sooho Jung, Keunho Park

Abstract:

In this paper, we describe an image post-processing method to enhance the performance of the parcel segmentation method using deep learning-based aerial images conducted in previous studies. The study results were evaluated using a confusion matrix, IoU, Precision, Recall, and F1-Score. In the case of the confusion matrix, it was observed that the false positive value, which is the result of misclassification, was greatly reduced as a result of image post-processing. The average IoU was 0.9688 in the image post-processing, which is higher than the deep learning result of 0.8362, and the F1-Score was also 0.9822 in the image post-processing, which was higher than the deep learning result of 0.8850. As a result of the experiment, it was found that the proposed technique positively complements the deep learning results in segmenting the parcel of interest.

Keywords: aerial image, image process, machine vision, open field smart farm, segmentation

Procedia PDF Downloads 46

Authors: Saba Rahimi, Ozan Oktay, Javier Alvarez-Valle, Sujeeth Bharadwaj

Abstract:

Successful application of deep learning in medical image analysis necessitates unprecedented amounts of labeled training data. Unlike conventional 2D applications, radiological images can be three-dimensional (e.g., CT, MRI), consisting of many instances within each image. The problem is exacerbated when expert annotations are required for effective pixel-wise labeling, which incurs exorbitant labeling effort and cost. Active learning is an established research domain that aims to reduce labeling workload by prioritizing a subset of informative unlabeled examples to annotate. Our contribution is a cost-effective approach for U-Net 3D models that uses Monte Carlo sampling to analyze pixel-wise uncertainty. Experiments on the AAPM 2017 lung CT segmentation challenge dataset show that our proposed framework can achieve promising segmentation results by using only 42% of the training data.

Keywords: image segmentation, active learning, convolutional neural network, 3D U-Net

Procedia PDF Downloads 119

Authors: Yagya Raj Pandeya, Joonwhoan Lee

Abstract:

The dynamic behavior of music and video makes it difficult to evaluate musical instrument playing in a video by computer system. Any television or film video clip with music information are rich sources for analyzing musical instruments using modern machine learning technologies. In this research, we integrate the audio and video information sources using convolutional neural network (CNN) and pass network learned features through recurrent neural network (RNN) to preserve the dynamic behaviors of audio and video. We use different pre-trained CNN for music and video feature extraction and then fine tune each model. The music network use 2D convolutional network and video network use 3D convolution (C3D). Finally, we concatenate each music and video feature by preserving the time varying features. The long short term memory (LSTM) network is used for long-term dynamic feature characterization and then use late fusion with generalized mean. The proposed network performs better performance to recognize the musical instrument using audio-video multimodal neural network.

Keywords: multimodal, 3D convolution, music-video feature extraction, generalized mean

Procedia PDF Downloads 186

Authors: Ahmedou Moulaye Idriss, Tfeil Yahya, Tamas Ungi, Gabor Fichtinger

Abstract:

Image-guided therapy (IGT) plays a crucial role in minimally invasive procedures for liver interventions. Accurate segmentation of the liver and precise needle placement is essential for successful interventions such as needle aspiration. In this study, we propose a deep learning-based liver 3D slicer designed to enhance segmentation accuracy and facilitate needle aspiration procedures. The developed 3D slicer leverages state-of-the-art convolutional neural networks (CNNs) for automatic liver segmentation in medical images. The CNN model is trained on a diverse dataset of liver images obtained from various imaging modalities, including computed tomography (CT) and magnetic resonance imaging (MRI). The trained model demonstrates robust performance in accurately delineating liver boundaries, even in cases with anatomical variations and pathological conditions. Furthermore, the 3D slicer integrates advanced image registration techniques to ensure accurate alignment of preoperative images with real-time interventional imaging. This alignment enhances the precision of needle placement during aspiration procedures, minimizing the risk of complications and improving overall intervention outcomes. To validate the efficacy of the proposed deep learning-based 3D slicer, a comprehensive evaluation is conducted using a dataset of clinical cases. Quantitative metrics, including the Dice similarity coefficient and Hausdorff distance, are employed to assess the accuracy of liver segmentation. Additionally, the performance of the 3D slicer in guiding needle aspiration procedures is evaluated through simulated and clinical interventions. Preliminary results demonstrate the effectiveness of the developed 3D slicer in achieving accurate liver segmentation and guiding needle aspiration procedures with high precision. The integration of deep learning techniques into the IGT workflow shows great promise for enhancing the efficiency and safety of liver interventions, ultimately contributing to improved patient outcomes.

Keywords: deep learning, liver segmentation, 3D slicer, image guided therapy, needle aspiration

Procedia PDF Downloads 13

Authors: Harry J. Witchel, James Ackah, Carlos P. Santos, Nachiappan Chockalingam, Carina E. I. Westling

Abstract:

Introduction: Proxemics is the study of how people perceive and use space. It is commonly proposed that when people like or engage with a person/object, they will move slightly closer to it, often quite subtly and subconsciously. Music videos are known to add entertainment value to a pop song. Our hypothesis was that by adding appropriately matched video to a pop song, it would lead to a net approach of the head to the monitor screen compared to simply listening to an audio-only version of the song. Methods: We presented to 27 participants (ages 21.00 ± 2.89, 15 female) seated in front of 47.5 x 27 cm monitor two musical stimuli in a counterbalanced order; all stimuli were based on music videos by the band OK Go: Here It Goes Again (HIGA, boredom ratings (0-100) = 15.00 ± 4.76, mean ± SEM, standard-error-of-the-mean) and Do What You Want (DWYW, boredom ratings = 23.93 ± 5.98), which did not differ in boredom elicited (P = 0.21, rank-sum test). Each participant experienced each song only once, and one song (counterbalanced) as audio-only versus the other song as a music video. The movement was measured by video-tracking using Kinovea 0.8, based on recording from a lateral aspect; before beginning, each participant had a reflective motion tracking marker placed on the outer canthus of the left eye. Analysis of the Kinovea X-Y coordinate output in comma-separated-variables format was performed in Matlab, as were non-parametric statistical tests. Results: We found that the audio-only stimuli (combined for both HIGA and DWYW, mean ± SEM, 35.71 ± 5.36) were significantly more boring than the music video versions (19.46 ± 3.83, P = 0.0066 Wilcoxon Signed Rank Test (WSRT), Cohen's d = 0.658, N = 28). We also found that participants' heads moved around twice as much during the audio-only versions (speed = 0.590 ± 0.095 mm/sec) compared to the video versions (0.301 ± 0.063 mm/sec, P = 0.00077, WSRT). However, the participants' mean head-to-screen distances were not detectably smaller (i.e. head closer to the screen) during the music videos (74.4 ± 1.8 cm) compared to the audio-only stimuli (73.9 ± 1.8 cm, P = 0.37, WSRT). If anything, during the audio-only condition, they were slightly closer. Interestingly, the ranges of the head-to-screen distances were smaller during the music video (8.6 ± 1.4 cm) compared to the audio-only (12.9 ± 1.7 cm, P = 0.0057, WSRT), the standard deviations were also smaller (P = 0.0027, WSRT), and their heads were held 7 mm higher (video 116.1 ± 0.8 vs. audio-only 116.8 ± 0.8 cm above floor, P = 0.049, WSRT). Discussion: As predicted, sitting and listening to experimenter-selected pop music was more boring than when the music was accompanied by a matched, professionally-made video. However, we did not find that the proxemics of the situation led to approaching the screen. Instead, adding video led to efforts to control the head to a more central and upright viewing position and to suppress head fidgeting.

Keywords: boredom, engagement, music videos, posture, proxemics

Procedia PDF Downloads 139

Authors: Ana Londral, Burcu Demiray, Marcus Cheetham

Abstract:

Speech recording is a methodology used in many different studies related to cognitive and behaviour research. Modern advances in digital equipment brought the possibility of continuously recording hours of speech in naturalistic environments and building rich sets of sound files. Speech analysis can then extract from these files multiple features for different scopes of research in Language and Communication. However, tools for analysing a large set of sound files and automatically extract relevant features from these files are often inaccessible to researchers that are not familiar with programming languages. Manual analysis is a common alternative, with a high time and efficiency cost. In the analysis of long sound files, the first step is the voice segmentation, i.e. to detect and label segments containing speech. We present a comprehensive methodology aiming to support researchers on voice segmentation, as the first step for data analysis of a big set of sound files. Praat, an open source software, is suggested as a tool to run a voice detection algorithm, label segments and files and extract other quantitative features on a structure of folders containing a large number of sound files. We present the validation of our methodology with a set of 5000 sound files that were collected in the daily life of a group of voluntary participants with age over 65. A smartphone device was used to collect sound using the Electronically Activated Recorder (EAR): an app programmed to record 30-second sound samples that were randomly distributed throughout the day. Results demonstrated that automatic segmentation and labelling of files containing speech segments was 74% faster when compared to a manual analysis performed with two independent coders. Furthermore, the methodology presented allows manual adjustments of voiced segments with visualisation of the sound signal and the automatic extraction of quantitative information on speech. In conclusion, we propose a comprehensive methodology for voice segmentation, to be used by researchers that have to work with large sets of sound files and are not familiar with programming tools.

Keywords: automatic speech analysis, behavior analysis, naturalistic environments, voice segmentation

Procedia PDF Downloads 258

Authors: Jiameng Wang, Yunfei Yin, Xiyu Deng

Abstract:

Parallel optimization is one of the important research topics of data mining at this stage. Taking Classification and Regression Tree (CART) parallelization as an example, this paper proposes a parallel data mining algorithm based on SSP-OGini-PCCP. Aiming at the problem of choosing the best CART segmentation point, this paper designs an S-SP model without data association; and in order to calculate the Gini index efficiently, a parallel OGini calculation method is designed. In addition, in order to improve the efficiency of the pruning algorithm, a synchronous PCCP pruning strategy is proposed in this paper. In this paper, the optimal segmentation calculation, Gini index calculation, and pruning algorithm are studied in depth. These are important components of parallel data mining. By constructing a distributed cluster simulation system based on SPARK, data mining methods based on SSP-OGini-PCCP are tested. Experimental results show that this method can increase the search efficiency of the best segmentation point by an average of 89%, increase the search efficiency of the Gini segmentation index by 3853%, and increase the pruning efficiency by 146% on average; and as the size of the data set increases, the performance of the algorithm remains stable, which meets the requirements of contemporary massive data processing.

Keywords: classification, Gini index, parallel data mining, pruning ahead

Procedia PDF Downloads 100

Authors: Plessen Mogens Graf

Abstract:

Within the context of multivariate time series segmentation, this paper proposes a method inspired by a posteriori optimal trading. After a normalization step, time series are treated channelwise as surrogate stock prices that can be traded optimally a posteriori in a virtual portfolio holding either stock or cash. Linear transaction costs are interpreted as hyperparameters for noise filtering. Trading signals, as well as trading signals obtained on the reversed time series, are used for unsupervised channelwise labeling before a consensus over all channels is reached that determines the final segmentation time instants. The method is model-free such that no model prescriptions for segments are made. Benefits of proposed approach include simplicity, computational efficiency, and adaptability to a wide range of different shapes of time series. Performance is demonstrated on synthetic and real-world data, including a large-scale dataset comprising a multivariate time series of dimension 1000 and length 2709. Proposed method is compared to a popular model-based bottom-up approach fitting piecewise affine models and to a recent model-based top-down approach fitting Gaussian models and found to be consistently faster while producing more intuitive results in the sense of segmenting time series at peaks and valleys.

Keywords: time series segmentation, model-free, trading-inspired, multivariate data

Procedia PDF Downloads 106

Authors: Alessio Mastrorillo, Giuseppe Silvi, Francesco Scagliola

Abstract:

In this work, we introduce an Ambisonics decoder with an implementation of the C-format, also called Super Stereo. This format is an alternative to conventional stereo and binaural decoding. Unlike those, this format conveys audio information from the horizontal plane and works with stereo speakers and headphones. The two C-format channels can also return a reconstructed planar B-format. This work provides an open-source implementation for this format. We implement an all-pass filter for signal quadrature, as required by the decoding equations. This filter works with six Biquads in a cascade configuration, with values for control frequency and quality factor discovered experimentally. The phase response of the filter delivers a small error in the 20-14.000Hz range. The decoder has been tested with audio sources up to 192kHz sample rate, returning pristine sound quality and detailed stereo image. It has been included in the Envelop for Live suite and is available as an open-source repository. This decoder has applications in Virtual Reality and 360° audio productions, music composition, and online streaming.

Keywords: ambisonics, UHJ, quadrature filter, virtual reality, Gerzon, decoder, stereo, binaural, biquad

Procedia PDF Downloads 64

Authors: Parisa Mansour

Abstract:

The coronavirus disease 2019 (COVID-19) was discovered and rapidly spread to various countries around the world since the end of 2019. Computed tomography (CT) images have been used as an important alternative to the time-consuming RT. PCR test. However, manual segmentation of CT images alone is a major challenge as the number of suspected cases increases. Thus, accurate and automatic segmentation of COVID-19 infections is urgently needed. Because the imaging features of the COVID-19 infection are different and similar to the background, existing medical image segmentation methods cannot achieve satisfactory performance. In this work, we try to build a deep convolutional neural network adapted for the segmentation of chest CT images with COVID-19 infections. First, we maintain a large and novel chest CT image database containing 165,667 annotated chest CT images from 861 patients with confirmed COVID-19. Inspired by the observation that the boundary of an infected lung can be improved by global intensity adjustment, we introduce a feature variable block into the proposed deep CNN, which adjusts the global features of features to segment the COVID-19 infection. The proposed PV array can effectively and adaptively improve the performance of functions in different cases. We combine features of different scales by proposing a progressive atrocious space pyramid fusion scheme to deal with advanced infection regions with various aspects and shapes. We conducted experiments on data collected in China and Germany and showed that the proposed deep CNN can effectively produce impressive performance.

Keywords: chest, COVID-19, chest Image, coronavirus, CT image, chest CT

Procedia PDF Downloads 27

Authors: Elleuch Imen

Abstract:

In this paper, it propose a new approach for learning Arabic as a foreign language via audio-visual translation, particularly subtitling. The approach consists of developing video sequences appropriate to different levels of learning (from A1 to C2) containing conversations, quizzes, games and others. Each video aims to achieve a specific objective, such as the correct pronunciation of Arabic words, the correct syntactic structuring of Arabic sentences, the recognition of the morphological characteristics of terms and the semantic understanding of statements. The subtitled videos obtained can be incorporated into different Arabic second language learning tools such as Moocs, websites, platforms, etc.

Keywords: arabic foreign language, learning, audio-visuel translation, subtitled videos

Procedia PDF Downloads 30