Search results for: temporal convolutional net

Authors: Nidhal Azawi

Abstract:

Colonoscopy procedure is widely used in the world to detect an abnormality. Early diagnosis can help to heal many patients. Because of the unavoidable artifacts that exist in colon images, doctors cannot detect a colon surface precisely. The purpose of this work is to improve the visual quality of colonoscopy videos to provide better information for physicians by removing some artifacts. This work complements a series of work consisting of three previously published papers. In this paper, Optic flow is used for motion compensation, and then consecutive images are aligned/registered to integrate some information to create a new image that has or reveals more information than the original one. Colon images have been classified into informative and noninformative images by using a deep neural network. Then, two different strategies were used to treat informative and noninformative images. Informative images were treated by using Lucas Kanade (LK) with an adaptive temporal mean/median filter, whereas noninformative images are treated by using Lucas Kanade with a derivative of Gaussian (LKDOG) with adaptive temporal median images. A comparison result showed that this work achieved better results than that results in the state- of- the- art strategies for the same degraded colon images data set, which consists of 1000 images. The new proposed algorithm reduced the error alignment by about a factor of 0.3 with a 100% successfully image alignment ratio. In conclusion, this algorithm achieved better results than the state-of-the-art approaches in case of enhancing the informative images as shown in the results section; also, it succeeded to convert the non-informative images that have very few details/no details because of the blurriness/out of focus or because of the specular highlight dominate significant amount of an image to informative images.

Keywords: optic flow, colonoscopy, artifacts, spatial temporal filter

Procedia PDF Downloads 91

Authors: Tomas Premoli, Sareh Rowlands

Abstract:

In this study, we investigate the impact of various convolutional neural network (CNN) architectures on the accuracy of diagnosing Alzheimer’s disease (AD) using patient MRI scans. Alzheimer’s disease is a debilitating neurodegenerative disorder that affects millions worldwide. Early, accurate, and non-invasive diagnostic methods are required for providing optimal care and symptom management. Deep learning techniques, particularly CNNs, have shown great promise in enhancing this diagnostic process. We aim to contribute to the ongoing research in this field by comparing the effectiveness of different CNN architectures and providing insights for future studies. Our methodology involved preprocessing MRI data, implementing multiple CNN architectures, and evaluating the performance of each model. We employed intensity normalization, linear registration, and skull stripping for our preprocessing. The selected architectures included VGG, ResNet, and DenseNet models, all implemented using the Keras library. We employed transfer learning and trained models from scratch to compare their effectiveness. Our findings demonstrated significant differences in performance among the tested architectures, with DenseNet201 achieving the highest accuracy of 86.4%. Transfer learning proved to be helpful in improving model performance. We also identified potential areas for future research, such as experimenting with other architectures, optimizing hyperparameters, and employing fine-tuning strategies. By providing a comprehensive analysis of the selected CNN architectures, we offer a solid foundation for future research in Alzheimer’s disease diagnosis using deep learning techniques. Our study highlights the potential of CNNs as a valuable diagnostic tool and emphasizes the importance of ongoing research to develop more accurate and effective models.

Keywords: Alzheimer’s disease, convolutional neural networks, deep learning, medical imaging, MRI

Procedia PDF Downloads 45

Authors: Daria N. Podvigina, Tatiana V. Chernigovskaya

Abstract:

We have studied the temporal characteristics of bistable perception of the stimuli of two types: one involves alterations in a perceived depth and another one has an ambiguous content. We used the Necker lattice and lines of shadowed circles ambiguously perceived either as spheres or holes as stimuli of the first type. The Winson figure (the Eskimo/Indian picture) was a stimulus of the second type. We have analyzed how often the reversals occurred (reversal rate) and for how long each of the two interpretations, or percepts, was observed during one presentation (stability durations). For all three ambiguous images the reversal rate and the stability durations had similar values, which provide another evidence for a significant role of top-down processes in multistable perception.

Keywords: multistable perception, perceived depth, reversal rate, top-down processes

Procedia PDF Downloads 555

Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan

Abstract:

The most aggressive form of brain tumor is called glioma. Glioma is kind of tumor that arises from glial tissue of the brain and occurs quite often. A fully automatic 2D-CNN model for brain tumor segmentation is presented in this paper. We performed pre-processing steps to remove noise and intensity variances using N4ITK and standard intensity correction, respectively. We used Keras open-source library with Theano as backend for fast implementation of CNN model. In addition, we used BRATS 2015 MRI dataset to evaluate our proposed model. Furthermore, we have used SimpleITK open-source library in our proposed model to analyze images. Moreover, we have extracted random 2D patches for proposed 2D-CNN model for efficient brain segmentation. Extracting 2D patched instead of 3D due to less dimensional information present in 2D which helps us in reducing computational time. Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.77 for complete, 0.76 for core, 0.77 for enhanced tumor regions. However, these results are comparable with methods already implemented 2D CNN architecture.

Keywords: brain tumor segmentation, convolutional neural networks, deep learning, LGG

Procedia PDF Downloads 158

Authors: Bright E. Owusu, Nittaya McNeil

Abstract:

Rainfall is one of the most essential quantities in meteorology and hydrology. It has important impacts on people’s daily life and excess or inadequate of it could bring tremendous losses in economy and cause fatalities. Population increase around the globe tends to have a corresponding increase in settlement and industrialization. Some countries are affected by flood and drought occasionally due to climate change, which disrupt most of the daily activities. Knowledge of trends in spatial and temporal rainfall variability and their physical explanations would be beneficial in climate change assessment and to determine erosivity. This study describes the spatial-temporal variability of daily rainfall in Australia and their corresponding long-term trend during 1950-2013. The spatial patterns were investigated by using exploratory factor analysis and the long term trend in rainfall time series were determined by linear regression, Mann-Kendall rank statistics and the Sen’s slope test. The exploratory factor analysis explained most of the variations in the data and grouped Australia into eight distinct rainfall regions with different rainfall patterns. Significant increasing trends in annual rainfall were observed in the northern regions of Australia. However, the northeastern part was the wettest of all the eight rainfall regions.

Keywords: climate change, explanatory factor analysis, Mann-Kendall and Sen’s slope test, rainfall.

Procedia PDF Downloads 324

Authors: Xinyuan Chen, Mohd Nizam Husen, Zhongmei Zhou, Gongde Guo, Wei Gao

Abstract:

Temporal Knowledge Graph Completion (TKGC) has attracted increasing attention for its enormous value; however, existing models lack capabilities to capture both local interactions and global dependencies simultaneously with evolutionary dynamics, while the latest achievements in convolutions and Transformers haven't been employed in this area. What’s more, periodic patterns in TKGs haven’t been fully explored either. To this end, a multi-stage hybrid architecture with convolution-backed Transformers is introduced in TKGC tasks for the first time combining the Hawkes process to model evolving event sequences in a continuous-time domain. In addition, the seasonal-trend decomposition is adopted to identify periodic patterns. Experiments on six public datasets are conducted to verify model effectiveness against state-of-the-art (SOTA) methods. An extensive ablation study is carried out accordingly to evaluate architecture variants as well as the contributions of independent components in addition, paving the way for further potential exploitation. Besides complexity analysis, input sensitivity and safety challenges are also thoroughly discussed for comprehensiveness with novel methods.

Keywords: temporal knowledge graph completion, convolution, transformer, Hawkes process, periodicity

Procedia PDF Downloads 51

Authors: Justin Reverdi, Sixin Zhang, Saïd Aoues, Fabrice Gamboa, Serge Gratton, Thomas Pellegrini

Abstract:

In vapor cycle systems, the mass flow sensor plays a key role for different monitoring and control purposes. However, physical sensors can be inaccurate, heavy, cumbersome, expensive, or highly sensitive to vibrations, which is especially problematic when embedded into an aircraft. The conception of a virtual sensor, based on other standard sensors, is a good alternative. This paper has two main objectives. Firstly, a data-driven model using a convolutional neural network is proposed to estimate the mass flow of the compressor. We show that it significantly outperforms the standard polynomial regression model (thermodynamic maps) in terms of the standard MSE metric and engineer performance metrics. Secondly, a semi-automatic segmentation method is proposed to compute the engineer performance metrics for real datasets, as the standard MSE metric may pose risks in analyzing the dynamic behavior of vapor cycle systems.

Keywords: deep learning, convolutional neural network, vapor cycle system, virtual sensor

Procedia PDF Downloads 26

Authors: Daniel Dahis, Haim Azhari

Abstract:

Noninvasive image-guided intracranial treatments using high intensity focused ultrasound (HIFU) are on the course of translation into clinical applications. They include, among others, tumor ablation, hyperthermia, and blood-brain-barrier (BBB) penetration. Since many of these procedures are associated with local temperature elevation, thermal monitoring is essential. MRI constitutes an imaging method with high spatial resolution and thermal mapping capacity. It is the currently leading modality for temperature guidance, commonly under the name MRgHIFU (magnetic-resonance guided HIFU). Nevertheless, MRI is a very expensive non-portable modality which jeopardizes its accessibility. Ultrasonic thermal monitoring, on the other hand, could provide a modular, cost-effective alternative with higher temporal resolution and accessibility. In order to assess the feasibility of ultrasonic brain thermal monitoring, this study investigated the usage of brain tissue attenuation coefficient (AC) temporal changes as potential estimators of thermal changes. Newton's law of cooling describes a temporal exponential decay behavior for the temperature of a heated object immersed in a relatively cold surrounding. Similarly, in the case of cerebral HIFU treatments, the temperature in the region of interest, i.e., focal zone, is suggested to follow the same law. Thus, it was hypothesized that the AC of the irradiated tissue may follow a temporal exponential behavior during cool down regime. Three ex-vivo bovine brain tissue specimens were inserted into plastic containers along with four thermocouple probes in each sample. The containers were placed inside a specially built ultrasonic tomograph and scanned at room temperature. The corresponding pixel-averaged AC was acquired for each specimen and used as a reference. Subsequently, the containers were placed in a beaker containing hot water and gradually heated to about 45ᵒC. They were then repeatedly rescanned during cool down using ultrasonic through-transmission raster trajectory until reaching about 30ᵒC. From the obtained images, the normalized AC and its temporal derivative as a function of temperature and time were registered. The results have demonstrated high correlation (R² > 0.92) between both the brain AC and its temporal derivative to temperature. This indicates the validity of the hypothesis and the possibility of obtaining brain tissue temperature estimation from the temporal AC thermal changes. It is important to note that each brain yielded different AC values and slopes. This implies that a calibration step is required for each specimen. Thus, for a practical acoustic monitoring of the brain, two steps are suggested. The first step consists of simply measuring the AC at normal body temperature. The second step entails measuring the AC after small temperature elevation. In face of the urging need for a more accessible thermal monitoring technique for brain treatments, the proposed methodology enables a cost-effective high temporal resolution acoustical temperature estimation during HIFU treatments.

Keywords: attenuation coefficient, brain, HIFU, image-guidance, temperature

Procedia PDF Downloads 139

Authors: Luke McGowan, David Pickernell, Martini Battisti

Abstract:

In this paper, Adversity Triggers were explored through the lens of Organisational Resilience. Adversity Triggers are contextualized by temporal factors, thus, naturally aligning to Resilience literature. Resilience has been chosen as the theoretical framework as risk management approaches are often not geared towards providing meaningful responses to high-impact, low-probability events. Adversity Triggers and Organisational Resilience both consider temporal factors which enabled investigation of each phase of recovery. A systematic literature was employed to assess previous literature and define further areas of research. The systematic literature review method was chosen to catalogue and identify gaps in current literature.

Keywords: adversity triggers, crisis, extreme events, organisational resilience, resilience

Procedia PDF Downloads 123

Authors: Elsayed A. Issawy, Anwar H. Radwan

Abstract:

Recent crustal deformations studies in Egypt are applied on the most active areas with relation to seismic activity. Temporal gravity variations in parallel with the geodetic technique (GPS) were used to monitor recent crustal movements in Egypt since 1997. The non-tidal gravity changes were constrained by the vertical component of surface movements derived from the GPS observations. The gravity changes were used to understand the surface tectonics and geodynamic modelling of the Greater Cairo region after the occurrence of an earthquake of 1992. It was found that there is a certain relation showed by increasing of gravity values before the main seismic activity. As example, relative considerable increase of gravity values was noticed for the network between the epochs of 2000 and 2004. Otherwise, the temporal gravity variations were reported a considerable decrease in gravity values between the two campaigns of 2004 and 2007 for the same stations. This behaviour could explain by compressive deformation and strain build-up stage before the South western Cairo earthquake (July 31, 2005 with magnitude of 4.3) and the stress release stage occurred after the main event. The geodetic measurements showed that, the estimated horizontal velocities for almost of points are 5.5 mm/year in approximately NW direction.

Keywords: temporal gravity variations, geodynamics, greater Cairo, recent crustal movements, earthquakes

Procedia PDF Downloads 344

Authors: Ryan S. Johnson, Raeesa Moolla

Abstract:

Airports are known air pollution hotspots due to the variety of fuel driven activities that take place within the confines of them. As such, people working within airports are particularly vulnerable to exposure of hazardous air pollutants, including hundreds of aromatic hydrocarbons, and more specifically a group of compounds known as BTEX (viz. benzene, toluene, ethyl-benzene and xylenes). These compounds have been identified as being harmful to human and environmental health. Through the use of passive and active sampling methods, the spatial and temporal variability of benzene, toluene, ethyl-benzene and xylene concentrations within the international airport was investigated. Two sampling campaigns were conducted. In order to quantify the temporal variability of concentrations within the airport, an active sampling strategy using the Synspec Spectras Gas Chromatography 955 instrument was used. Furthermore, a passive sampling campaign, using Radiello Passive Samplers was used to quantify the spatial variability of these compounds. In addition, meteorological factors are known to affect the dispersal and dilution of pollution. Thus a Davis Pro-Weather 2 station was utilised in order to measure in situ weather parameters (viz. wind speed, wind direction and temperature). Results indicated that toluene varied on a daily, temporal scale considerably more than other concentrations. Toluene further exhibited a strong correlation with regards to the meteorological parameters, inferring that toluene was affected by these parameters to a greater degree than the other pollutants. The passive sampling campaign revealed BTEXtotal concentrations ranged between 12.95 – 124.04 µg m-3. From the results obtained it is clear that benzene, toluene, ethyl-benzene and xylene concentrations are heterogeneously spatially dispersed within the airport. Due to the slow wind speeds recorded over the passive sampling campaign (1.13 m s-1.), the hotspots were located close to the main concentration sources. The most significant hotspot was located over the main apron of the airport. It is recommended that further, extensive investigations into the seasonality of hazardous air pollutants at the airport is necessary in order for sound conclusions to be made about the temporal and spatial distribution of benzene, toluene, ethyl-benzene and xylene concentrations within the airport.

Keywords: airport, air pollution hotspot, BTEX concentrations, meteorology

Procedia PDF Downloads 172

Authors: Ankit Kamboj, Shikha Talwar, Nilesh Powar

Abstract:

To successfully deliver our products in the market, the employees need to be in a safe environment, especially in an industrial and manufacturing environment. The consequences of delinquency in wearing safety glasses while working in industrial plants could be high risk to employees, hence the need to develop a real-time automatic detection system which detects the persons (violators) not wearing safety glasses. In this study a convolutional neural network (CNN) algorithm called faster region based CNN (Faster RCNN) with rotated bounding box has been used for detecting safety glasses on persons; the algorithm has an advantage of detecting safety glasses with different orientation angles on the persons. The proposed method of rotational bounding boxes with a convolutional neural network first detects a person from the images, and then the method detects whether the person is wearing safety glasses or not. The video data is captured at the entrance of restricted zones of the industrial environment (manufacturing plant), which is further converted into images at 2 frames per second. In the first step, the CNN with pre-trained weights on COCO dataset is used for person detection where the detections are cropped as images. Then the safety goggles are labelled on the cropped images using the image labelling tool called roLabelImg, which is used to annotate the ground truth values of rotated objects more accurately, and the annotations obtained are further modified to depict four coordinates of the rectangular bounding box. Next, the faster RCNN with rotated bounding box is used to detect safety goggles, which is then compared with traditional bounding box faster RCNN in terms of detection accuracy (average precision), which shows the effectiveness of the proposed method for detection of rotatory objects. The deep learning benchmarking is done on a Dell workstation with a 16GB Nvidia GPU.

Keywords: CNN, deep learning, faster RCNN, roLabelImg rotated bounding box, safety goggle detection

Procedia PDF Downloads 111

Authors: Mohammad Tahsin Karimi Nezhad, Zaynab Shadbahr, Ali Gholami

Abstract:

In recent years many attempts have been made to assess groundwater contamination by nitrates worldwide. The assessment of spatial and temporal variations of physico-chemical parameters of water is necessary to mange water quality. The objectives of the study were to evaluate spatial variability and temporal changes of hydrochemical factors by water sampling from 24 wells in the Shoush City suburb. The analysis was conducted for the whole area and for different land use and geological classes. In addition, nitrate concentration variability with descriptive parameters such as sampling depth, dissolved oxygen, and on ground nitrogen loadings was also investigated The results showed that nitrate concentrations did not exceed the standard limit (50 mg/l). EC of water samples, ranged from 900 to 1200 µs/cm, TDS from 775 to 830 mg/l and pH from 5.6 to 9.

Keywords: groundwater, GIS, water quality, Iran

Procedia PDF Downloads 407

Authors: Dena Alhaloul

Abstract:

The Middle East and North Africa (MENA) region is characterized by diverse cultures, economies, and social structures. Fertility rates in MENA have seen significant changes over time, with variations among countries and subregions. Understanding fertility patterns in this region is essential due to its impact on demographic dynamics, healthcare, labor markets, and social policies. Rising or declining fertility rates have far-reaching consequences for the region's socioeconomic development. The main thrust of this study is to comprehensively examine fertility rates in the Middle East and North Africa (MENA) region. It aims to understand the distribution, determinants, and temporal trends of fertility rates in MENA countries. The study seeks to provide insights into the factors influencing fertility decisions, assess how fertility rates have evolved over time, and potentially develop statistical models to characterize these trends. As for the methodology of the study, the study uses descriptive statistics to summarize and visualize fertility rate data. It also uses regression analyses to identify determinants of fertility rates as well as statistical modeling to characterize temporal trends in fertility rates. The conclusion of this study The research will contribute to a deeper understanding of fertility dynamics in the MENA region, shedding light on the distribution of fertility rates, their determinants, and historical trends.

Keywords: fertility, distribution, modeling, regression

Procedia PDF Downloads 44

Authors: Fidel Lòpez Saca, Carlos Avilés-Cruz, Miguel Magos-Rivera, José Antonio Lara-Chávez

Abstract:

Automated object recognition and identification systems are widely used throughout the world, particularly in assembly lines, where they perform quality control and automatic part selection tasks. This article presents the design and implementation of an object recognition system in an assembly line. The proposed shapes-color recognition system is based on deep learning theory in a specially designed convolutional network architecture. The used methodology involve stages such as: image capturing, color filtering, location of object mass centers, horizontal and vertical object boundaries, and object clipping. Once the objects are cut out, they are sent to a convolutional neural network, which automatically identifies the type of figure. The identification system works in real-time. The implementation was done on a Raspberry Pi 3 system and on a Jetson-Nano device. The proposal is used in an assembly course of bachelor’s degree in industrial engineering. The results presented include studying the efficiency of the recognition and processing time.

Keywords: deep-learning, image classification, image identification, industrial engineering.

Procedia PDF Downloads 133

Authors: Zhuo Job Chen, Kevin Nute

Abstract:

The work reported examines potential linkages between spatial and temporal prospects, and more specifically, between variations in the spatial depth and foreground obstruction of window views, and observers’ sense of connection to the future. It was found that external views from indoor spaces were strongly associated with a sense of the future, that partially obstructing such a view with foreground objects significantly reduced its association with the future, and replacing it with a pictorial representation of the same scene (with no real actual depth) removed most of its temporal association. A lesser change in the spatial depth of the view, however, had no apparent effect on association with the future. While the role of spatial depth has still to be confirmed, the results suggest that spatial prospects directly affect temporal ones. The word “prospect” typifies the overlapping of the spatial and temporal in most human languages. It originated in classical times as a purely spatial term, but in the 16th century took on the additional temporal implication of an imagined view ahead, of the future. The psychological notion of prospection, then, has its distant origins in a spatial analogue. While it is not yet proven that space directly structures our processing of time at a physiological level, it is generally agreed that it commonly does so conceptually. The mental representation of possible futures has been a central part of human survival as a species (Boyer, 2008; Suddendorf & Corballis, 2007). A sense of the future seems critical not only practically, but also psychologically. It has been suggested, for example, that lack of a positive image of the future may be an important contributing cause of depression (Beck, 1974; Seligman, 2016). Most people in the developed world now spend more than 90% of their lives indoors. So any direct link between external views and temporal prospects could have important implications for both human well-being and building design. We found that the ability to see what lies in front of us spatially was strongly associated with a sense of what lies ahead temporally. Partial obstruction of a view was found to significantly reduce that sense connection to the future. Replacing a view with a flat pictorial representation of the same scene removed almost all of its connection with the future, but changing the spatial depth of a real view appeared to have no significant effect. While foreground obstructions were found to reduce subjects’ sense of connection to the future, they increased their sense of refuge and security. Consistent with Prospect and Refuge theory, an ideal environment, then, would seem to be one in which we can “see without being seen” (Lorenz, 1952), specifically one that conceals us frontally from others, without restricting our own view. It is suggested that these optimal conditions might be translated architecturally as screens, the apertures of which are large enough for a building occupant to see through unobstructed from close by, but small enough to conceal them from the view of someone looking from a distance outside.

Keywords: foreground obstructions, prospection, spatial depth, window views

Procedia PDF Downloads 96

Authors: M. A. Hernández-Ceballos, M. Marín-Ferrer, G. Cinelli, L. De Felice, T. Tollefsen, E. Nweke, P. V. Tognoli, S. Vanzo, M. De Cort

Abstract:

This study analyses the latitudinal impact of the spatial and temporal distribution on the cosmogenic isotope 7Be in surface air along Europe. The long-term database of the 6 sampling sites (Ivalo, Helsinki, Berlin, Freiburg, Sevilla and La Laguna), that regularly provide data to the Radioactivity Environmental Monitoring (REM) network managed by the Joint Research Centre (JRC) in Ispra, were used. The selection of the stations was performed attending to different factors, such as 1) heterogeneity in terms of latitude and altitude, and 2) long database coverage. The combination of these two parameters ensures a high degree of representativeness of the results. In the later, the temporal coverage varies between stations, being used in the present study sampling stations with a database more or less continuously from 1984 to 2011. The mean values of 7Be activity concentration presented a spatial distribution value ranging from 2.0 ± 0.9 mBq/m3 (Ivalo, north) to 4.8 ± 1.5 mBq/m3 (La Laguna, south). An increasing gradient with latitude was observed from the north to the south, 0.06 mBq/m3. However, there was no correlation with altitude, since all stations are sited within the atmospheric boundary layer. The analyses of the data indicated a dynamic range of 7Be activity for solar cycle and phase (maximum or minimum), having been observed different impact on stations according to their location. The results indicated a significant seasonal behavior, with the maximum concentrations occurring in the summer and minimum in the winter, although with differences in the values reached and in the month registered. Due to the large heterogeneity in the temporal pattern with which the individual radionuclide analyses were performed in each station, the 7Be monthly index was calculated to normalize the measurements and perform the direct comparison of monthly evolution among stations. Different intensity and evolution of the mean monthly index were observed. The knowledge of the spatial and temporal distribution of this natural radionuclide in the atmosphere is a key parameter for modeling studies of atmospheric processes, which are important phenomena to be taken into account in the case of a nuclear accident.

Keywords: Berilium-7, latitudinal impact in Europe, seasonal and monthly variability, solar cycle

Procedia PDF Downloads 318

Authors: R. Ramesh, K. K. Shivaraman

Abstract:

The main aim of this paper is to handle the energy requirement in an efficient manner by merging the advanced digital communication and control technologies for smart grid applications. In order to reduce user home load during peak load hours, utility applies several incentives such as real-time pricing, time of use, demand response for residential customer through smart meter. However, this method provides inconvenience in the sense that user needs to respond manually to prices that vary in real time. To overcome these inconvenience, this paper proposes a convolutional neural network (CNN) with k-means clustering machine learning model which have ability to forecast energy requirement in short term, i.e., hour of the day or day of the week. By integrating our proposed technique with home energy management based on Bluetooth low energy provides predicted value to user for scheduling appliance in advanced. This paper describes detail about CNN configuration and k-means clustering algorithm for short-term energy forecasting.

Keywords: convolutional neural network, fuzzy logic, k-means clustering approach, smart home energy management

Procedia PDF Downloads 281

Authors: Daleesha M. Viswanathan, Sumam Mary Idicula

Abstract:

Dynamic hand gestures are an intrinsic component in sign language communication. Extracting spatial temporal features of the hand gesture trajectory plays an important role in a dynamic gesture recognition system. Finding a discrete feature descriptor for the motion trajectory based on the orientation feature is the main concern of this paper. Kalman filter algorithm and Hidden Markov Models (HMM) models are incorporated with this recognition system for hand trajectory tracking and for spatial temporal classification, respectively.

Keywords: orientation features, discrete feature vector, HMM., Indian sign language

Procedia PDF Downloads 342

Authors: Philipp Ruf, Massiwa Chabbi, Christoph Reich, Djaffar Ould-Abdeslam

Abstract:

In recent years, convolutional neural networks (CNN) have demonstrated high performance in image analysis, but oftentimes, there is only structured data available regarding a specific problem. By interpreting structured data as images, CNNs can effectively learn and extract valuable insights from tabular data, leading to improved predictive accuracy and uncovering hidden patterns that may not be apparent in traditional structured data analysis. In applying a single neural network for analyzing multimodal data, e.g., both structured and unstructured information, significant advantages in terms of time complexity and energy efficiency can be achieved. Converting structured data into images and merging them with existing visual material offers a promising solution for applying CNN in multimodal datasets, as they often occur in a medical context. By employing suitable preprocessing techniques, structured data is transformed into image representations, where the respective features are expressed as different formations of colors and shapes. In an additional step, these representations are fused with existing images to incorporate both types of information. This final image is finally analyzed using a CNN.

Keywords: CNN, image processing, tabular data, mixed dataset, data transformation, multimodal fusion

Procedia PDF Downloads 86

Authors: Hira Lal Gope, Hidekazu Fukai

Abstract:

The aim of this study is to develop a system which can identify and sort peaberries automatically at low cost for coffee producers in developing countries. In this paper, the focus is on the classification of peaberries and normal coffee beans using image processing and machine learning techniques. The peaberry is not bad and not a normal bean. The peaberry is born in an only single seed, relatively round seed from a coffee cherry instead of the usual flat-sided pair of beans. It has another value and flavor. To make the taste of the coffee better, it is necessary to separate the peaberry and normal bean before green coffee beans roasting. Otherwise, the taste of total beans will be mixed, and it will be bad. In roaster procedure time, all the beans shape, size, and weight must be unique; otherwise, the larger bean will take more time for roasting inside. The peaberry has a different size and different shape even though they have the same weight as normal beans. The peaberry roasts slower than other normal beans. Therefore, neither technique provides a good option to select the peaberries. Defect beans, e.g., sour, broken, black, and fade bean, are easy to check and pick up manually by hand. On the other hand, the peaberry pick up is very difficult even for trained specialists because the shape and color of the peaberry are similar to normal beans. In this study, we use image processing and machine learning techniques to discriminate the normal and peaberry bean as a part of the sorting system. As the first step, we applied Deep Convolutional Neural Networks (CNN) and Support Vector Machine (SVM) as machine learning techniques to discriminate the peaberry and normal bean. As a result, better performance was obtained with CNN than with SVM for the discrimination of the peaberry. The trained artificial neural network with high performance CPU and GPU in this work will be simply installed into the inexpensive and low in calculation Raspberry Pi system. We assume that this system will be used in under developed countries. The study evaluates and compares the feasibility of the methods in terms of accuracy of classification and processing speed.

Keywords: convolutional neural networks, coffee bean, peaberry, sorting, support vector machine

Procedia PDF Downloads 118

Authors: Rima Hleiss, Elie Bitar, Mahmoud Hassan, Mohamad Khalil

Abstract:

A comprehensive study of object recognition in the human brain requires combining both spatial and temporal analysis of brain activity. Here, we are mainly interested in three issues: the time perception of visual objects, the ability of discrimination between two particular categories (objects vs. animals), and the possibility to identify a particular spatial representation of visual objects. Our experiment consisted of acquiring dense electroencephalographic (EEG) signals during a picture-naming task comprising a set of objects and animals’ images. These EEG responses were recorded from nine participants. In order to determine the time perception of the presented visual stimulus, we analyzed the Event Related Potentials (ERPs) derived from the recorded EEG signals. The analysis of these signals showed that the brain perceives animals and objects with different time instants. Concerning the discrimination of the two categories, the support vector machine (SVM) was applied on the instantaneous EEG (excellent temporal resolution: on the order of millisecond) to categorize the visual stimuli into two different classes. The spatial differences between the evoked responses of the two categories were also investigated. The results showed a variation of the neural activity with the properties of the visual input. Results showed also the existence of a spatial pattern of electrodes over particular regions of the scalp in correspondence to their responses to the visual inputs.

Keywords: brain activity, categorization, dense EEG, evoked responses, spatio-temporal analysis, SVM, time perception

Procedia PDF Downloads 394

Authors: Trupti Chavan, Ramachandra Guda, Kameshwar Rao

Abstract:

The lack of data is a pain problem in medical image analysis using a convolutional neural network (CNN). This work uses various spatial transformation techniques to address the medical image augmentation issue for knee detection and localization using an enhanced single shot detector (SSD) network. The spatial transforms like a negative, histogram equalization, power law, sharpening, averaging, gaussian blurring, etc. help to generate more samples, serve as pre-processing methods, and highlight the features of interest. The experimentation is done on the OpenKnee dataset which is a collection of knee images from the openly available online sources. The CNN called enhanced single shot detector (SSD) is utilized for the detection and localization of the knee joint from a given X-ray image. It is an enhanced version of the famous SSD network and is modified in such a way that it will reduce the number of prediction boxes at the output side. It consists of a classification network (VGGNET) and an auxiliary detection network. The performance is measured in mean average precision (mAP), and 99.96% mAP is achieved using the proposed enhanced SSD with spatial transformations. It is also seen that the localization boundary is comparatively more refined and closer to the ground truth in spatial augmentation and gives better detection and localization of knee joints.

Keywords: data augmentation, enhanced SSD, knee detection and localization, medical image analysis, openKnee, Spatial transformations

Procedia PDF Downloads 130

Authors: Harsh Sadawarti, Kamal Malik

Abstract:

Humans are the unique and the most powerful creature on this planet just because of the high level of intelligence gifted by nature. Computational Intelligence is highly influenced by the term natural intelligence, neurosciences and mathematics. To deal with the in-depth study of computational intelligence and to utilize it in real-life applications, it is quite important to understand its simulation with the human brain. In this paper, the three important parts, Frontal Lobe, Occipital Lobe and Parietal Lobe of the human brain, are compared with the ANN(Artificial Neural Network), CNN(Convolutional Neural network), and RNN(Recurrent Neural Network), respectively. Intelligent computational systems are created by combining deductive reasoning, logical concepts and high-level algorithms with the simulation and study of the human brain. Human brain is a combination of Physiology, Psychology, emotions, calculations and many other parameters which are of utmost importance that determines the overall intelligence. To create intelligent algorithms, smart machines and to simulate the human brain in an effective manner, it is quite important to have an insight into the human brain and the basic concepts of biological neurosciences.

Keywords: computational intelligence, neurosciences, convolutional neural network, recurrent neural network, artificial neural network, frontal lobe, occipital lobe, parietal lobe

Procedia PDF Downloads 84

Authors: Sonia Perez-Gamboa, Qingquan Sun, Yan Zhang

Abstract:

Non-intrusive sensor-based human activity recognition (HAR) is utilized in a spectrum of applications, including fitness tracking devices, gaming, health care monitoring, and smartphone applications. Deep learning models such as convolutional neural networks (CNNs) and long short term memory (LSTM) recurrent neural networks (RNNs) provide a way to achieve HAR accurately and effectively. In this paper, we design a multi-layer hybrid architecture with CNN and LSTM and explore a variety of multi-layer combinations. Based on the exploration, we present a lightweight, hybrid, and multi-layer model, which can improve the recognition performance by integrating local features and scale-invariant with dependencies of activities. The experimental results demonstrate the efficacy of the proposed model, which can achieve a 94.7% activity recognition rate on a benchmark human activity dataset. This model outperforms traditional machine learning and other deep learning methods. Additionally, our implementation achieves a balance between recognition rate and training time consumption.

Keywords: deep learning, LSTM, CNN, human activity recognition, inertial sensor

Procedia PDF Downloads 118

Authors: Muhammad Atif, Cang Yan

Abstract:

The enhancement of low-light images is a significant area of study aimed at enhancing the quality of captured images in challenging lighting environments. Recently, methods based on convolutional neural networks (CNN) have gained prominence as they offer state-of-the-art performance. However, many approaches based on CNN rely on increasing the size and complexity of the neural network. In this study, we propose an alternative method for improving low-light images using an autoencoder-based multiscale knowledge transfer model. Our method leverages the power of three autoencoders, where the encoders of the first two autoencoders are directly connected to the decoder of the third autoencoder. Additionally, the decoder of the first two autoencoders is connected to the encoder of the third autoencoder. This architecture enables effective knowledge transfer, allowing the third autoencoder to learn and benefit from the enhanced knowledge extracted by the first two autoencoders. We further integrate the proposed model into the PIX to PIX GAN framework. By integrating our proposed model as the generator in the GAN framework, we aim to produce enhanced images that not only exhibit improved visual quality but also possess a more authentic and realistic appearance. These experimental results, both qualitative and quantitative, show that our method is better than the state-of-the-art methodologies.

Keywords: low light image enhancement, deep learning, convolutional neural network, image processing

Procedia PDF Downloads 35

Authors: Salam Khalifa, Naveed Ahmed

Abstract:

We present a new method to reconstruct a temporally coherent 3D animation from single or multi-view RGB-D video data using unbiased feature point sampling. Given RGB-D video data, in form of a 3D point cloud sequence, our method first extracts feature points using both color and depth information. In the subsequent steps, these feature points are used to match two 3D point clouds in consecutive frames independent of their resolution. Our new motion vectors based dynamic alignment method then fully reconstruct a spatio-temporally coherent 3D animation. We perform extensive quantitative validation using novel error functions to analyze the results. We show that despite the limiting factors of temporal and spatial noise associated to RGB-D data, it is possible to extract temporal coherence to faithfully reconstruct a temporally coherent 3D animation from RGB-D video data.

Keywords: 3D video, 3D animation, RGB-D video, temporally coherent 3D animation

Procedia PDF Downloads 342

Authors: E. U. Dampage, Ovindi D. Bandara, Vinushi S. Waraketiya, Samitha S. R. De Silva, Yasiru S. Gunarathne

Abstract:

The knowledge of water inflow figures is paramount in decision making on the allocation for consumption for numerous purposes; irrigation, hydropower, domestic and industrial usage, and flood control. The understanding of how reservoir inflows are affected by different climatic and hydrological conditions is crucial to enable effective water management and downstream flood control. In this research, we propose a method using a Long Short Term Memory (LSTM) Artificial Neural Network (ANN) to assist the aforesaid decision-making process. The Kotmale reservoir, which is the uppermost reservoir in the Mahaweli reservoir complex in Sri Lanka, was used as the test bed for this research. The ANN uses the runoff in the Kotmale reservoir catchment area and the effect of Sea Surface Temperatures (SST) to make a forecast for seven days ahead. Three types of ANN are tested; Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), and LSTM. The extensive field trials and validation endeavors found that the LSTM ANN provides superior performance in the aspects of accuracy and latency.

Keywords: convolutional neural network, CNN, inflow, long short-term memory, LSTM, multi-layer perceptron, MLP, neural network

Procedia PDF Downloads 121

Authors: Amin Nezarat, Naeime Seifadini

Abstract:

Trip mode selection is a behavioral characteristic of passengers with immense importance for travel demand analysis, transportation planning, and traffic management. Identification of trip mode distribution will allow transportation authorities to adopt appropriate strategies to reduce travel time, traffic and air pollution. The majority of existing trip mode inference models operate based on human selected features and traditional machine learning algorithms. However, human selected features are sensitive to changes in traffic and environmental conditions and susceptible to personal biases, which can make them inefficient. One way to overcome these problems is to use neural networks capable of extracting high-level features from raw input. In this study, the convolutional neural network (CNN) architecture is used to predict the trip mode distribution based on raw GPS trajectory data. The key innovation of this paper is the design of the layout of the input layer of CNN as well as normalization operation, in a way that is not only compatible with the CNN architecture but can also represent the fundamental features of motion including speed, acceleration, jerk, and Bearing rate. The highest prediction accuracy achieved with the proposed configuration for the convolutional neural network with batch normalization is 85.26%.

Keywords: predicting, deep learning, neural network, urban trip

Procedia PDF Downloads 106

Authors: Nasser S. Shebka

Abstract:

Fuzzy logic is used in complex adaptive systems where classical tools of representing knowledge are unproductive. Nevertheless, the incorporation of fuzzy logic, as it’s the case with all artificial intelligence tools, raised some inconsistencies and limitations in dealing with increased complexity systems and rules that apply to real-life situations and hinders the ability of the inference process of such systems, but it also faces some inconsistencies between inferences generated fuzzy rules of complex or imprecise knowledge-based systems. The use of fuzzy logic enhanced the capability of knowledge representation in such applications that requires fuzzy representation of truth values or similar multi-value constant parameters derived from multi-valued logic, which set the basis for the three t-norms and their based connectives which are actually continuous functions and any other continuous t-norm can be described as an ordinal sum of these three basic ones. However, some of the attempts to solve this dilemma were an alteration to fuzzy logic by means of non-monotonic logic, which is used to deal with the defeasible inference of expert systems reasoning, for example, to allow for inference retraction upon additional data. However, even the introduction of non-monotonic fuzzy reasoning faces a major issue of conflict resolution for which many principles were introduced, such as; the specificity principle and the weakest link principle. The aim of our work is to improve the logical representation and functional modelling of AI systems by presenting a method of resolving existing and potential rule conflicts by representing temporal modalities within defeasible inference rule-based systems. Our paper investigates the possibility of resolving fuzzy rules conflict in a non-monotonic fuzzy reasoning-based system by introducing temporal modalities and Kripke's general weak modal logic operators in order to expand its knowledge representation capabilities by means of flexibility in classifying newly generated rules, and hence, resolving potential conflicts between these fuzzy rules. We were able to address the aforementioned problem of our investigation by restructuring the inference process of the fuzzy rule-based system. This is achieved by using time-branching temporal logic in combination with restricted first-order logic quantifiers, as well as propositional logic to represent classical temporal modality operators. The resulting findings not only enhance the flexibility of complex rule-base systems inference process but contributes to the fundamental methods of building rule bases in such a manner that will allow for a wider range of applicable real-life situations derived from a quantitative and qualitative knowledge representational perspective.

Keywords: fuzzy rule-based systems, fuzzy tense inference, intelligent systems, temporal modalities

Procedia PDF Downloads 63