Your browser doesn't support javascript.
Mostrar: 20 | 50 | 100
Resultados 1 - 5 de 5
Mais filtros

Base de dados
Intervalo de ano de publicação
Sensors (Basel) ; 21(17)2021 Aug 30.
Artigo em Inglês | MEDLINE | ID: mdl-34502738


In the field of computer vision, object detection consists of automatically finding objects in images by giving their positions. The most common fields of application are safety systems (pedestrian detection, identification of behavior) and control systems. Another important application is head/person detection, which is the primary material for road safety, rescue, surveillance, etc. In this study, we developed a new approach based on two parallel Deeplapv3+ to improve the performance of the person detection system. For the implementation of our semantic segmentation model, a working methodology with two types of ground truths extracted from the bounding boxes given by the original ground truths was established. The approach has been implemented in our two private datasets as well as in a public dataset. To show the performance of the proposed system, a comparative analysis was carried out on two deep learning semantic segmentation state-of-art models: SegNet and U-Net. By achieving 99.14% of global accuracy, the result demonstrated that the developed strategy could be an efficient way to build a deep neural network model for semantic segmentation. This strategy can be used, not only for the detection of the human head but also be applied in several semantic segmentation applications.

Pedestres , Semântica , Humanos , Processamento de Imagem Assistida por Computador , Redes Neurais de Computação
Materials (Basel) ; 14(7)2021 Mar 30.
Artigo em Inglês | MEDLINE | ID: mdl-33808116


Polycarbonate-carbon nanotube (PC-CNT) conductive composites containing CNT concentration covering 0.25-4.5 wt.% were prepared by melt blending extrusion. The alternating current (AC) conductivity of the composites has been investigated. The percolation threshold of the PC-CNT composites was theoretically determined using the classical theory of percolation followed by numerical analysis, quantifying the conductivity of PC-CNT at the critical volume CNT concentration. Different theoretical models like Bueche, McCullough and Mamunya have been applied to predict the AC conductivity of the composites using a hyperparameter optimization method. Through multiple series of the hyperparameter optimization process, it was found that McCullough and Mamunya theoretical models for electrical conductivity fit remarkably with our experimental results; the degree of chain branching and the aspect ratio are estimated to be 0.91 and 167 according to these models. The development of a new model based on a modified Sohi model is in good agreement with our data, with a coefficient of determination R2=0.922 for an optimized design model. The conductivity is correlated to the electromagnetic absorption (EM) index showing a fine fit with Steffen-Boltzmann (SB) model, indicating the ultimate CNTs volume concentration for microwave absorption at the studied frequency range.

Micromachines (Basel) ; 11(8)2020 Aug 15.
Artigo em Inglês | MEDLINE | ID: mdl-32824164


Electronic devices that transmit, distribute, or utilize electrical energy create electromagnetic interference (EMI) that can lead to malfunctioning and degradation of electronic devices. EMI shielding materials block the unwanted electromagnetic waves from reaching the target material. EMI issues can be solved by using a new family of building blocks constituted of polymer and nanofillers. The electromagnetic absorption index of this material is calculated by measuring the "S-parameters". In this article, we investigated the use of artificial intelligence (AI) in the EMI shielding field by developing a new system based on a multilayer perceptron neural network designed to predict the electromagnetic absorption of polycarbonate-carbon nanotubes composites films. The proposed system included 15 different multilayer perception (MLP) networks; each network was specialized to predict the absorption value of a specific category sample. The selection of appropriate networks was done automatically, using an independent block. Optimization of the hyper-parameters using hold-out validation was required to ensure the best results. To evaluate the performance of our system, we calculated the similarity error, precision accuracy, and calculation time. The results obtained over our database showed clearly that the system provided a very good result with an average accuracy of 99.7997%, with an overall average calculation time of 0.01295 s. The composite based on polycarbonate-5 wt.% carbon nanotube was found to be the ultimate absorber over microwave range according to Rozanov formalism.

Pan Afr Med J ; 24: 322, 2016.
Artigo em Francês | MEDLINE | ID: mdl-28154677


Malignant otitis externa is also referred to as skull base osteomyelitis. Pseudomonas aeruginosa is the most common offending pathogen. However, fungal origin is not uncommon. 80-year-old patient having persistent left earache since two months despite adequate treatment. Otologic examination showed signs of inflammation in the auricular pavilion, ear canal stenosis with granulomas and purulent otorrhea. CT scan showed a filled otomastoid, extensive inflammatory process affecting the pre-auricular and retro-auricular tissues and lysis of the tympanic bulla. Given the lack of improvement, mycological examination was performed which revealed the presence of Candida Albicans. Cases of malignant otitis externa caused by Candida albicans are rarely reported. Fungal origin should be suspected in patients who have negative bacteriological samples and no improvement despite adequate antibiotic treatment. It should be confirmed by sometimes multiple mycological samples. Malignant otitis externa caused by Candida albicans is a rare potentially mortal infection.

Candida albicans/isolamento & purificação , Candidíase/diagnóstico , Otite Externa/diagnóstico , Idoso de 80 Anos ou mais , Candidíase/microbiologia , Feminino , Humanos , Otite Externa/microbiologia , Otomicose/diagnóstico , Otomicose/microbiologia
Pan Afr Med J ; 24: 269, 2016.
Artigo em Francês | MEDLINE | ID: mdl-28154624


External auditory canal cholesteatoma is defined as an accumulation of keratin leading to osteolytic bony erosion. It is a rare or underdiagnosed otologic entity. Our study aims to illustrate one case of external auditory canal cholesteatoma. Hypertensive diabetic patient, 65 years of age, under therapy suffering from severe right earache, insomnia, hearing loss, small purulent otorrhea and right facial paralysis grade V since 3 months. Initially the diagnosis was malignant external otitis. The patient underwent antibiotic therapy without improvement. Otologic examination showed burgeoning ulcerative lesion of the posterior wall of the right external auditory meatus. Systematic lesion biopsy was performed and confirmed the diagnosis of cholesteatoma. The patient underwent open technique tympanoplasty. External auditory canal cholesteatoma is rare, clinical symptoms are not specific, coronal CT scan makes a major contribution to a positive diagnosis showing bony crater of external auditory meatus. Treatment depends on the extent of the lesions. It can be based on simple local cares or on open technique tympanoplasty. External auditory canal cholesteatoma may have multiple clinical aspects and lend to confusion with other pathologies of the external auditory meatus.

Colesteatoma/diagnóstico , Meato Acústico Externo/patologia , Otopatias/diagnóstico , Timpanoplastia/métodos , Idoso , Biópsia/métodos , Colesteatoma/patologia , Colesteatoma/cirurgia , Otopatias/patologia , Otopatias/cirurgia , Humanos , Masculino , Otite Externa/diagnóstico