Deep Learning-Based Fire Detection for Enhanced Safety Systems

Wasit Journal of Pure sciences, Год журнала: 2023, Номер 2(4), С. 45 - 55

Опубликована: Дек. 30, 2023

Fire detection systems are a critical aspect of modern safety and security systems, playing pivotal role in safeguarding lives property against the destructive force fires. Rapid accurate identification fire incidents is essential for timely response mitigation efforts. Traditional methods have made substantial advancements, but with advent computer vision technologies, field has witnessed transformative shift. This paper presents method using deep convolutional neural network (CNN) models. approach used transfer learning by employing two pre-trained CNN models from ImageNet dataset: VGG (Visual Geometry Group) InceptionV3 to extract valuable features input images. Then, these extracted serve as machine (ML) classifier, namely Softmax classifier. The activation function computes probability distribution assign class probabilities discriminating between types images: non-fire. Experimental results showed that proposed successfully detected areas achieved seamless classification performance compared other current methods.

Язык: Английский

---

COVID-19 Vaccine Tracking Geospatial Application using GIS and GPS

TEM Journal, Год журнала: 2023, Номер unknown, С. 603 - 613

Опубликована: Май 29, 2023

One of the problems in addressing COVID -19 epidemic is that coverage vaccination data each area cannot be immediately displayed. In addition, COVID-19 presented do not include spatial data, which means used for decision making to address problem. Although are through digital platforms. However, a current limitation platforms manipulate have supported real-time manipulation, processing, or visualization. As result, distribution vaccinated individuals tracked. Therefore, this research developed geospatial application processing and visualization using Geographic Information System (GIS) Global Positioning (GPS) plan support by people officials so operations can conducted efficiently collection time reduced. The software develop platform includes PHP, MySQL, Google Maps API Leaflet. results show monitor track real COVID-19.

Язык: Английский

---

Speech Recognition Algorithms based Cough Recognition System

International Journal of Online and Biomedical Engineering (iJOE), Год журнала: 2023, Номер 19(12), С. 49 - 61

Опубликована: Авг. 31, 2023

This paper introduces an innovative technique for creating a cough detection system that relies on speech recognition algorithms. The strategy utilizes the Kaldi platform, which is open source and incorporates hybrid of Gaussian Mixture Model-based Hidden Markov Models (GMM-HMM) through straightforward monophone training model. Additionally, study examines effectiveness two different feature extraction approaches, Mel Frequency Cepstral Coefficient (MFCC) Perceptual Linear Prediction (PLP). proposed can function as collection tool gathering natural spontaneous data from conversations or continuous speech. also compares CMU Sphinx4 toolkits, concluding Kaldi’s use GMM-HMM outperforms Sphinx4.

Язык: Английский

---

Discriminative Approach Lung Diseases and COVID-19 from Chest X-Ray Images Using Convolutional Neural Networks: A Promising Approach for Accurate Diagnosis

International Journal of Online and Biomedical Engineering (iJOE), Год журнала: 2023, Номер 19(14), С. 131 - 141

Опубликована: Окт. 11, 2023

Medical imaging treatment is one of the best-known computer science disciplines. It can be used to detect presence several diseases such as skin cancer and brain tumors, since arrival coronavirus (COVID-19), this technique has been alleviate heavy burden placed on all health institutions personnel, given high rate spread virus in population. One problems encountered diagnosing people suspected having contracted COVID-19 difficulty distinguishing symptoms due from those other influenza, they are similar. This paper proposes a new approach between lung by analyzing chest x-ray images using convolutional neural network (CNN) architecture. To achieve this, pre-processing was carried out dataset histogram equalization, then we trained two sub-datasets Train et Test, first training phase second model validation phase. Then CNN architecture composed convolution layers fully connected deployed train our model. Finally, evaluated different metrics: confusion matrix receiver operating characteristic. The simulation results recorded satisfactory, with an accuracy 96.27%.

Язык: Английский

---

Multi-label Classification Technique of Chest X-Rays Image Based Cardiomegaly Disease Prediction

Опубликована: Март 20, 2023

One of the most common techniques used in detecting serious life-threatening diseases is chest X-Ray radiography, through which datasets X-ray images are collected. Among heart-related that can be detected using this technique Cardiomegaly. However, image-based identification considers a time-consuming process and requires radiologists with high skills to interpret analyze these accurately diagnose pathologies, especially for difficult cases cannot interpreted by naked eyes humans where one image may have more than pathology. In context, solve above problems. paper, we deal problems designing efficient architecture two automated classification models based on transfer learning convolution neural network DenseNet121 as feature engineering. These architectures were constructed from backbone model followed proposed deep consists global average pooling layer, layers, an output layer. The first was designed multi-label predict 8 types diseases, thus layer contains neurons sigmoid activation function, while second focused binary cardiomyopathy so neuron. Two custom functions multi label suitable its task, loss calculation accuracy function. performance implemented CheXpert dataset evaluated terms area under curve (AUC). results show achieved AUC score 90% obtained 83% consider promising results. addition, web application interface produced work contributed practicality applicable it examined practical clinical prove generalization models, testing good realistic.

Язык: Английский

---