Abstract
In the age of smart cities, it is envisioned that most processes within the smart city context will be smart and automated. This includes smart houses, smart kitchens, etc. within this context, a need will arise for Smart nursery rooms. Within the smart nursery concept, the infant needs will need to be fulfilled automatically, in addition, to infant monitoring and safety. The motivation of this work is to design a smart cradle system for a smart nursery room that automates the functions of the cradle based on the infant’s sounds. Therefore, in this paper, we propose an infant sound classification technique based on the Support Vector Classifier (SVC) with Radial Basis Function (RBF) kernel using 18 extracted features of infant sounds. The proposed technique has been compared with two SVC kernel function, linear, and poly, as well as other classification algorithms including Decision Tree, Random Forest, and Gaussian Naive Bayes. As a result of comparing the confusion matrix, recall, F1 Score, accuracies, and precision values of various applied machine learning algorithms over-extracted features. SVC using RBF kernel function was found to be the most efficient model with an average accuracy of more than 96%. The proposed system outperforms all other systems proposed in the previous literature.
-
Author Keywords
- Baby crying Detection,
- Smart Nursery,
- Smart Cities,
- Support Vector Classifier,
- Decision Tree,
- Random Forest
-
IEEE Keywords
- Feature extraction,
- Pediatrics,
- Smart cities,
- Support vector machines,
- Machine learning,
- Classification algorithms,
- Forestry
Introduction
In the age of the Internet of Things (IoT), Big Data, Cloud Computing, Artificial Intelligence, Robotics, and other new technologies, the concept of SMART CITIES is gaining more interest. Smart Cities are cities visualized to be fully automated to serve the residents. Simple tasks such as cooking, taking out the trash, cleaning, and others will all be automated. Part of the smart cities’ visualization is smart houses in which each room in the house and even the surroundings of the house including safety and irrigation is automated. Family houses that include a nursery room will require the nursery room to be smart as well. The aspects of smart rooms especially a nursery room is very complex and a lot of research has to be done in automating the processes that monitor, serve, and ensure the safety of infants. Integrating the new technologies in order to design systems that automate and provide services to humans is an area of research that spans over multiple disciplines. Smart Cities [1], Smart Medicine [2], Smart Education [3], and other disciplines are all jumping on the wagon of IoT, machine learning, AI, and new technologies in proposing state of the art systems to better human life. In this paper, we propose the use of machine learning to design a smart cradle for a smart nursery that will be part of the smart cities. Nowadays, parents go out of their way to check whether their babies are well or not especially during late-night hours. For instance, an employed parent might be required to complete some job-related tasks, or he/she might need to do some household chores while their baby is sleeping. Therefore, having a smart cradle system in such situations is an essential need to provide safety for the infant and peace of mind for the parents. The aim of this research is to design a smart cradle system within a smart nursery for smart cities based on machine algorithms to detect whether the infant is sleeping or crying and take the necessary automated actions. In the past, in order to detect whether a baby is crying or not, people used baby monitors to detect the baby’s sound frequency characteristics. This monitor performs the signal amplification, timing, shaping and filtering when there is input; and during the testing, the output, as a result, will prove if the detected sound is a cry or not. However, with the rapid development of technology, Machine Learning is utilized nowadays to perform the automatic detection and classification of baby sounds including crying. Machine learning (ML) is when the computer systems study the algorithms and arithmetic models in order to implement a particular assignment efficiently by using inference and patterns instead of being directed by a specific instruction [4]. In ML, there are many ways in which the machine can learn. It could be Supervised, Unsupervised, and Reinforcement Learning. When the machine is supervised, it means it is learning from a supervisor that acts as a teacher to provide labeled data to be helpful in training. Unlike Unsupervised learning, the machine is fed with an unknown dataset and it identifies the pattern of its data by plotting them into the x-y axis to see in the clusters to identify features. In Reinforcement learning, the learning is working based on the principle of feedback. For instance, we give the system unlabeled data to identify its characteristics as an output; when the system gives a wrong output, then we give negative feedback to the machine by providing the accurate output. Therefore, the machine will learn from the feedback that will be provided to it. However, Supervised learning, includes classification and regression as two main categories [5]. Feature-based Machine learning as well as deep learning is an area of research where various input data formats (voice, images, waveforms, etc.) are used with the intent of increasing the automatic detection and classification and reduce human intervention [6-8]. The motivation of this research is to eventually design a smart nursery room for smart houses in smart cities. However, due to the complexity of the problem and design, we will start in this paper by designing a smart cradle for the smart nursery room. The aim of this paper is to design a smart automated cradle that can detect and classify infant sound and take the necessary action based on the classification output. In this paper, the focus will be on the classification of baby environmental sounds into 4 different classes. The rest of the paper is organized as follows: Section 2 details the literature review, section 3 shows the details of the dataset used in this research, section 4 explains the proposed method, section 5 shows all the results with comments and section 6 concludes the research. Section 7 lists all the references used in this study.
Conclusion
Designing a full smart nursery for smart cities is a complex task and will require research in machine learning, AI, smart systems, etc. In this paper, we propose the design of one component in the smart nursery that of a smart cradle using machine learning and novel features. The smart cradle is intended to help parents ensure the safety and comfort of their infants. The complete smart cradle design is envisioned to contain features of automatically switching the cradle on in certain situations, notify parents in other situations, and other smart functionalities. All these decisions and which action to take depends on the detection and classification of the baby sounds. For the detection and classification of the baby sounds, we proposed in this paper the use of novel features and through experiments determined that SVC using RBF is the classifier of choice in the system that achieved the most accurate results among other classifiers. 18 extracted features were used. The system is compared with four other supervised classifiers, on the dataset leads to the conclusion that illustrates the SVC with RBF kernel achieved the best accuracies of more than 96% using proposed features. For future work, we intend to continue the design of the smart nursery with all its components as well as continue the work on the smart cradle to achieve better accuracy with the classification of more than the 4 categories of sounds in the dataset used.
About KSRA
The Kavian Scientific Research Association (KSRA) is a non-profit research organization to provide research / educational services in December 2013. The members of the community had formed a virtual group on the Viber social network. The core of the Kavian Scientific Association was formed with these members as founders. These individuals, led by Professor Siavosh Kaviani, decided to launch a scientific / research association with an emphasis on education.
KSRA research association, as a non-profit research firm, is committed to providing research services in the field of knowledge. The main beneficiaries of this association are public or private knowledge-based companies, students, researchers, researchers, professors, universities, and industrial and semi-industrial centers around the world.
Our main services Based on Education for all Spectrum people in the world. We want to make an integration between researches and educations. We believe education is the main right of Human beings. So our services should be concentrated on inclusive education.
The KSRA team partners with local under-served communities around the world to improve the access to and quality of knowledge based on education, amplify and augment learning programs where they exist, and create new opportunities for e-learning where traditional education systems are lacking or non-existent.
FULL Paper PDF file:
Smart Nursery for Smart Cities: Infant Sound Classification Based on Novel Features and Support Vector ClassifierBibliography
author
Year
2020
Title
Smart Nursery for Smart Cities: Infant Sound Classification Based on Novel Features and Support Vector Classifier
Publish in
2020 7th International Conference on Electrical and Electronics Engineering (ICEEE), Antalya, Turkey, 2020, pp. 47-52,
Doi
10.1109/ICEEE49618.2020.9102574.
PDF reference and original file: Click here
Somayeh Nosrati was born in 1982 in Tehran. She holds a Master's degree in artificial intelligence from Khatam University of Tehran.
Professor Siavosh Kaviani was born in 1961 in Tehran. He had a professorship. He holds a Ph.D. in Software Engineering from the QL University of Software Development Methodology and an honorary Ph.D. from the University of Chelsea.
Nasim Gazerani was born in 1983 in Arak. She holds a Master's degree in Software Engineering from UM University of Malaysia.
