Wyniki wyszukiwania - BazTech

1

The Role of Faster R-CNN Algorithm in the Internet of Things to Detect Mask Wearing: The Endemic Preparations

Nasution Marah Doly, Al-Khowarizmi, Rahmat Romi Fadillah, Lubis Arif Ridho, Lubis Muharman

International Journal of Electronics and Telecommunications

|

2023

|

Vol. 69, No. 4

691--696

EN

Faster R-CNN is an algorithm development that continuously starts from CNN then R-CNN and Faster R-CNN. The development of the algorithm is needed to test whether the heuristic algorithm has optimal provisions. Broadly speaking, faster R-CNN is included in algorithms that are able to solve neural network and machine learning problems to detect a moving object. One of the moving objects in the current phenomenon is the use of masks. Where various countries in the world have issued endemic orations after the Covid 19 pandemic occurred. Detection tool has been prepared that has been tested at the mandatory mask door, namely for mask users. In this paper, the role of the Faster R-CNN algorithm has been carried out to detect masks poured on Internet of Thinks (IoT) devices to automatically open doors for standard mask users. From the results received that testing on the detection of moving mask objects when used reaches 100% optimal at a distance of 0.5 to 1 meter and 95% at a distance of 1.5 to 2 meters so that the process of sending detection signals to IoT devices can be carried out at a distance of 1 meter at the position mask users to automatic doors.

2

Remote sensing detection and resource utilisation of urban sewage sludge based on mobile edge computing

Hu Weijie, Zhou Youfei, Lu Junying, Sheng Jun, Jin Zechen

Ecological Chemistry and Engineering. S

|

2023

|

Vol. 30, nr 2

275--282

EN

Improper disposal of municipal sewage sludge poses a significant threat to effective environmental protection. With the continuous advancement of artificial intelligence technology and the Internet of Things (IoT), remote sensing detection technology is emerging as a promising research avenue to address this issue. However, the current state of real-time detection technology is inadequate, hindering comprehensive and stable monitoring operation. Additionally, the rational use of network resources remains suboptimal. To address this challenge, this study proposes a resource optimisation technology for the current insufficient intelligent monitoring system of urban sewage sludge. By leveraging IoT and wireless technology, water meter data can be collected with minimal earth construction compared to traditional PLC collection. This is followed by utilising Faster R-CNN to plan the network transmission of sewage remote sensing information resources. Finally, the architecture collection module’s scalability is enhanced by incorporating edge computing and reserving sensor ports to meet future plant expansion demands. The experiment demonstrates the significant potential of this technology in application and resource optimisation. In actual parameter tracking tests, the proposed method effectively monitors sewage sludge, providing policy guidance and measure optimisation for relevant authorities, ultimately contributing to pollution-free urban development.

3

A new super resolution Faster R-CNN model based detection and classification of urine sediments

Avci Derya, Sert Eser, Dogantekin Esin, Yildirim Ozal, Tadeusiewicz Ryszard, Plawiak Pawel

Biocybernetics and Biomedical Engineering

|

2023

|

Vol. 43, no. 1

58--68

EN

The diagnosis of urinary tract infections and kidney diseases using urine microscopy images has gained significant attention of medical community in recent years. These images are usually created by physicians’ own rule of thumb manually. However, this manual urine sediment analysis is usually labor-intensive and time-consuming. In addition, even when physicians carefully examine an image, an erroneous cell recognition may occur due to some optical illusions. In order to achieve cell recognition in low-resolution urine microscopy images with a higher level of accuracy, a new super resolution Faster Region-based Convolutional Neural Network (Faster R-CNN) method is proposed. It aims to increase resolution in low-resolution urine microscopy images using self-similarity based single image super resolution which was used during the pre-processing. Denoising based Wiener filter and Discrete Wavelet Transform (DWT) are used to de-noise high resolution images, respectively, to increase the level of accuracy for image recognition. Finally, for the feature extraction and classification stages, AlexNet, VGFG16 and VGG19 based Faster R-CNN models are used for the recognition and detection of multi-class cells. The model yielded accuracy rates are 98.6%, 96.4% and 96.2% respectively.

4

Model Faster R-CNN uczony na syntetycznych obrazach

Łach Błażej, Łukasik Edyta

Journal of Computer Sciences Institute

|

2020

|

Vol. 17

401--404

PL

Uczenie maszynowe wymaga opisu danych przez człowieka. Opisywanie zbioru danych ręcznie jest bardzo czasochłonne. W artykule zbadano jak model uczył się na zdjęciach sztucznie wytworzonych, z jak najmniejszym udziałem człowieka przy opisywaniu danych. Sprawdzono jaki wpływ miało zastosowanie augmentacji i progresywnego rozmiaru zdjęcia przy treningu modelu na syntetycznym zbiorze. Model osiągnął nawet o 3,35% wyższą średnią precyzję na syntetycznym zbiorze danych przy zastosowaniu treningów z rosnącą rozdzielczością. Augmentacje poprawiły jakość detekcji na rzeczywistych zdjęciach. Wytwarzanie sztucznie danych treningowych ma duży wpływ na przyśpieszenie przygotowania treningów, ponieważ nie wymaga tak dużych nakładów ludzkich, jak klasyczne uczenie modeli z danymi opisanymi przez człowieka.

EN

Machine learning requires a human description of the data. The manual dataset description is very time consuming. In this article was examined how the model learns from artificially created images, with the least human participation in describing the data. It was checked how the model learned on artificially produced images with augmentations and progressive image size. The model has achieve up to 3.35 higher mean average precision on syntetic dataset in the training with increasing images resolution. Augmentations improved the quality of detection on real photos. The production of artificially generated training data has a great impact on the acceleration of prepare training, because it does not require as much human resources as normal learning process.