Object Recognition Using Convolutional Neural Networks

Menezes, Richardson Santiago Teles de; Magalhães, Rafael Marrocos; Maia, Helton

doi:10.5772/intechopen.89726

Cited by 9 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Gambar 1 Model Deteksi YOLO [12] Gambar 1 memperlihatkan arsitektur YOLO mirip dengan jaringan saraf konvolusi yang terinspirasi oleh model GoogLeNet untuk klasifikasi gambar. Lapisan awal jaringan pertama-tama mengekstrak fitur gambar, dan lapisan yang terhubung sepenuhnya memprediksi probabilitas dan koordinat keluaran.…”

Section: Yolounclassified

Perancangan Sistem Deteksi Pelanggaran Penggunaan Helm Dengan Metode Deep Learning Menggunakan Yolov5 Ultralytic

Batubara,

Hamonangan,

Arfan

et al. 2024

Transient

View full text Add to dashboard Cite

Penegakan aturan terkait penggunaan helm pada sepeda motor di lingkungan kampus menjadi upaya yang dilakukan untuk mengurangi dampak kecelakaan yang fatal, namun nyatanya masih banyak pengendara yang tidak sadar dan tidak taat akan hal ini. Pihak keamanan kampus yang terbatas dibanding dengan pelanggaran yang banyak terjadi telah menjadi keluhan dalam usaha menciptakan lingkungan kampus yaang aman. Meningkatnya teknologi machine learning yang semakin berkembang dan mudah diakses dapat diterapkan menjadi sistem deteksi pelanggaran dalam upaya membantu pihak keamanan dalam memonitoring keamanan berkendara. Sistem deteksi pelanggaran penggunaan helm dirancang menggunakan YOLOv5 Ultralytic dengan memanfaatkan metode deep learning yang didasari dengan algoritma convolutional neural network. YOLOv5 memberikan model dengan penekanan pada kecepatan inferensi yang baik, sehingga memungkinkan penggunaan dalam aplikasi real-time. Model akhir hasil training menggunakan yolov5 memiliki mean average precision (mAP) rata-rata sebesar 0,938 yang mengindikasikan model dapat melakukan deteksi objek dengan akurat dan konsisten. Model deteksi selanjutnya diintegrasikan menjadi sistem deteksi yang tersambung dengan database. Sistem deteksi ini akan menangkap dan mengirimkan gambar pelanggaran ke database. Hasil pengujian akhir sistem deteksi dengan melakukan evaluasi confusion matrix yang didapat menunjukkan akurasi sistem sebesar 98,5%. Gambar tersebut nantinya akan dimonitoring pada aplikasi android untuk membantu pihak keamanan dalam mengambil kebijakan yang tepat.

show abstract

Section: Yolounclassified

Perancangan Sistem Deteksi Pelanggaran Penggunaan Helm Dengan Metode Deep Learning Menggunakan Yolov5 Ultralytic

Batubara,

Hamonangan,

Arfan

et al. 2024

Transient

View full text Add to dashboard Cite

show abstract

“…In [18], Menezes et al discuss the latest object identification methods, such as you only look once, faster region convolutional neural network, and single-shot multibox detector. The authors, however, do not focus on the real-time food calorie estimation.…”

Section: Literature Reviewmentioning

confidence: 99%

Lightweight and Parameter-Optimized Real-Time Food Calorie Estimation from Images Using CNN-Based Approach

et al. 2022

View full text Add to dashboard Cite

Automated object identification has seen significant progress during the last decade with close to human-level accuracy, aided by deep learning methods. With the rapid rise of obesity and other lifestyle-related diseases worldwide, the availability of fast, automated, and reliable image-based food calorie estimation is becoming a necessity. With the help of a deep learning-based automated object identification system, it is possible to introduce accurate and intelligent solutions in the form of a mobile app. However, for these kind of applications, processing speed is an important concern as the images should be processed in real time. Although plenty of studies have been conducted that focus on food image detection-based calorie estimation, there is still a lack of an image-driven, lightweight, fast, and reliable food calorie estimation system. In this paper, we propose a method based on the parameter-optimized Convolution Neural Networks (CNN) for detecting food images of regular meals using a handheld camera. Once identification process of the food items are complete, the corresponding calories and nutritional facts can be calculated using prior knowledge about the food class. Through our findings, we demonstrate that our proposed approach ensures high accuracy and can significantly simplify the existing manual calorie estimation procedures by converting them into a real-time automated process.

show abstract

“…These systems primarily detect motion using image processing and background reduction to indicate human presence (Ahamad et al, 2020). Apart from this, several other detection algorithms exist, such as the pre-trained convolutional neural network (CNN) based models (Ahamad et al, 2020;Teles de Menezes et al, 2020). CNN-based models contribute to the automation filtering and feature extraction process of deep learning algorithms for object detection and classification (Teles de Menezes et al (2020).…”

Section: Introductionmentioning

confidence: 99%

“…Apart from this, several other detection algorithms exist, such as the pre-trained convolutional neural network (CNN) based models (Ahamad et al, 2020;Teles de Menezes et al, 2020). CNN-based models contribute to the automation filtering and feature extraction process of deep learning algorithms for object detection and classification (Teles de Menezes et al (2020). Several deep object detection algorithms such as You Only Look Once (YOLO), Single Shot Detector (SSD), Faster Region-based Convolutional Neural Network (Faster R-CNN), and Region-based Fully Convolutional Neural Network (R-FCN) were already studied for monitoring the adherence of the people to social distancing (Ahamad et al, 2020;Bhambani et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

Investigating the RSSI-based Distance Classification using Median Confidence Interval in a Multi-Device BLE Environment

VERANO,

SORONGON,

CRUZ

2023

EPSTEM

View full text Add to dashboard Cite

As the COVID-19 global pandemic occurred in 2020, there has been an increase in interest in using Received Signal Strength Indicator (RSSI) for strict implementations of social distancing. Bluetooth Low Energy (BLE) has been mainly used as an option for it is cost-effective and can be integrated with IoT for a wide range of applications. For a multiple-device BLE environment, the devices were calibrated, and parameters of the distance estimation were estimated using linear regression. For each device, RSSI values were acquired in various directions and distances. A confidence interval of 90% was used to create a prediction range for distance classification using the acquired RSSI. Using the range of median, BLE combinations at various distances and positions yield less than 50% of the data when classifying received RSSI values to distances of 1, 2, 3, and 4m. Irregularities from acquiring the RSSI value for various distances have been observed and can affect the classification of distance with the RSSI values. Further study is needed on other methods for the basis of interval range and minimization of irregularities.

show abstract

Object Recognition Using Convolutional Neural Networks

Cited by 9 publications

References 25 publications

Perancangan Sistem Deteksi Pelanggaran Penggunaan Helm Dengan Metode Deep Learning Menggunakan Yolov5 Ultralytic

Perancangan Sistem Deteksi Pelanggaran Penggunaan Helm Dengan Metode Deep Learning Menggunakan Yolov5 Ultralytic

Lightweight and Parameter-Optimized Real-Time Food Calorie Estimation from Images Using CNN-Based Approach

Investigating the RSSI-based Distance Classification using Median Confidence Interval in a Multi-Device BLE Environment

Contact Info

Product

Resources

About