<p class="Abstrak">Penggunaan ponsel sudah sangat erat dengan kehidupaan anak usia dini sehingga menimbulkan beberapa dampak negatif bagi anak usia dini terutama berkurangnya interaksi dengan dunia sekitarnya. Salah satu teknologi yang dapat dikembangkan pada ponsel adalah <em>computer vision. </em>Salah satu penggunaan <em>computer vision </em>adalah <em>object recognition</em> yang memberikan solusi untuk membantu mengenali objek.<em> </em>Pada penelitian ini dibangun sistem pengenalan objek benda di dalam rumah yang diaplikasikan pada ponsel yang diharapkan membantu anak usia dini mengenali benda disekitarnya. <em>MobileNet </em>merupakan salah satu <em>feature extraction</em> yang memiliki kinerja yang baik dan ringan digunakan pada perangkat ponsel. Arsitektur <em>MobileNet </em>terdiri dari <em>layer depthwise convolution </em>dan <em>layer pointwise </em><em>convolution </em>dalam mengekstraksi fitur<em>. </em>Percobaan ini juga menggunakan arsitektur <em>Single Shot Multibox Detector (SSD) </em>sebagai metode dalam mendeteksi objek<em>. Pre-trained model </em>dari dataset <em>COCO </em>digunakan pada eksperimen<em>,</em> untuk mengenali 20 jenis objek benda di dalam rumah. Dari hasil eksperimen, <em>MobileNetV2 </em>menghasilkan nilai <em>mean Average Precision (mAP)</em> yang lebih baik dibandingkan dengan <em>MobileNetV1 </em>dan<em> InceptionV2, </em>yaitu<em> </em>sebesar 99,34%.</p><p class="Abstrak"> </p><p class="Judul2"><strong><em>Abstract</em></strong></p><p class="Judul2"> <em>Mobile phone usage has been very close to early childhood life, so giving rise to some negative impact on early childhood, especially reduced interaction with the surrounding world. One of the technologies that can be developed on the cellphone is computer vision. One of the uses of computer vision is object recognition that provides solutions to help to recognize objects. This research builds a system for recognition objects inside in house that is developed on a cellphone that is expected to help early childhood recognize objects in the surrounding. MobileNet is one of feature extraction that has good performance and efficient use on a cellphone. MobileNet architecture consists of a depthwise convolution layer and pointwise convolution layer in extracting features. The experiment also uses the architecture of Single Shot Multibox Detector (SSD) as a method of detecting objects. We used MobileNet architecture as a pre-trained model that had previously been trained on COCO datasets, and implement transfer learning for 20 types of objects commonly found inside the house. The experimental result indicates that the mean Average Precision (mAP) of MobileNetV2 could exceed MobileNetV1 and InceptionV2 of 99.34%.</em></p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.