Solar dryer is typically used for agricultural purposes in Indonesia. There are many economically important crops requiring storage or drying under particular environmental conditions such as temperature and humidity. High temperatures inside solar dryer prevents the growth of microorganism, and quickly reduce moisture content from the substance. A hybrid solar dryer is generally considered to provide the most optimum solution, however solar panels may be expensive and they still only provide heat or energy in the daytime. Hence, we propose here a new kind of hybrid solar dryer for 24/7 optimum conditions for crops - enabled by recent advances in energy technologies as well as Industry 4.0. This study aims to create an efficient, affordable and a self-sufficient intelligent energy system that will be applied to agriculture for storage or drying purposes by measuring the energy needs for the optimal drying system. Therefore, it is crucial to estimate and assess the critical energy needs for such new systems in order to optimize and design such smart solar dryer (SSD) system especially for Indonesia’s agricultural needs. We use design experience of our industry partner (PT Impack Pratama Industri, Indonesia) who has been working extensively on such solar dryer dome (SDD) based on polycarbonate material (only solar irradiation, no other technologies) and theoretical framework based on first principles in thermodynamics to estimate and assess critical energy needs for such dome with all the smart technologies. The calculation was performed based on Mollier diagram and the result still a rough estimation of energy required.

Due to the explosive growth of the Web, the domain of Web personalization has gained great momentum both in the research and commercial areas. One of the most popular web personalization systems is recommender systems. In recommender systems choosing user information that can be used to profile users is very crucial for user profiling. In Web 2.0, one facility that can help users organize Web resources of their interest is user tagging systems. Exploring user tagging behavior provides a promising way for understanding users' information needs since tags are given directly by users. However, free and relatively uncontrolled vocabulary makes the user self-defined tags lack of standardization and semantic ambiguity. Also, the relationships among tags need to be explored since there are rich relationships among tags which could provide valuable information for us to better understand users. In this paper, we propose a novel approach for learning tag ontology based on the widely used lexical database WordNet for capturing the semantics and the structural relationships of tags. We present personalization strategies to disambiguate the semantics of tags by combining the opinion of WordNet lexicographers and users' tagging behavior together. To personalize further, clustering of users is performed to generate a more accurate ontology for a particular group of users. In order to evaluate the usefulness of the tag ontology, we use the tag ontology in a pilot tag recommendation experiment for improving the recommendation performance by exploiting the semantic information in the tag ontology. The initial result shows that the personalized information has improved the accuracy of the tag recommendation.

Many economically essential crops in Indonesia (such as coffee, tea, chocolate, or copra) require storage or drying under certain environmental conditions, especially temperature and humidity. The solar dryer dome, typically used for agricultural purposes in Indonesia, produces a sufficient amount of heat to increase the evaporation rate inside the dome and reduce the moisture content of the commodity. A hybrid solar dryer accompanied by a photovoltaic panel, fan, and ventilation system is generally suitable. The system can provide an optimum environment with minimum control. However, as the outdoor temperature and humidity change dramatically, such as at night time, more control is required. Based on Industry 4.0 technologies, we have developed a new kind of hybrid solar dryer that provides an optimum environment 24/7. The system, called Smart Dome 4.0, is an intelligent, low-cost, self-sufficient drying and storage system to support Indonesia Agriculture 4.0. The system has a local power generation unit to self-sustain the required energy and operate without connecting to the electricity grid. The system utilizes a machine-learning algorithm to predict the environmental condition and optimally uses self-generated electric power. The developed Smart Dome 4.0 technology is critical to producing a sustainable solar dome under drastic environmental dynamics.

The human face can be used for face recognition in order to increase the level of security of a safe deposit box because every person has his/her facial characteristics that have similarities with one another. One of the tasks for face recognition is to compare the face in real-time to the ones in the dataset so that the owner can be verified. This final project aims to implement face recognition based using Raspberry Pi to increase the level of security of the safe deposit box system design. This study uses the Raspberry Pi 3B+ because it has sufficient processing capabilities and has a few pre-built modules that make researching this less difficult. Raspberry Pi uses Linux as the operating system, which has access to a large number of libraries and applications compatible with it [1]. Of the many methods used for face detection, in this final project the Viola-Jones method is being used. From the result of this research, the success rate that was obtained is 60%. This number was obtained after 40 trials, the system was able to detect as much as 24 times [2]. The final results shows that the light intensity greatly affects the performance of the system. The light intensity of 8 Lux has an accuracy rate of 30%, while the 40 Lux has an accuracy rate of 90%.Wajah manusia dapat digunakan dalam pengenalan wajah untuk meningkatkan keamanan brankas karena setiap manusia memiliki fitur-fitur wajah yang berbeda-beda. Salah satu tugas dari pengenalan wajah adalah membandingkan wajah pada citra foto dengan wajah yang telah disimpan di dataset, agar identitas pemilik wajah dapat diketahui. Makalah ini mengimplementasikan pengenalan wajah berbasis Raspberry Pi pada sistem brankas untuk meningkatkan keamanan brankas. Penelitian ini menggunakan Raspberry Pi dikarenakan memiliki kemampuan pemrosesan yang cukup dan memiliki modul yang mempermudah implementasi. Raspberry Pi menggunakan Linux sebagai sistem operasi, yang memiliki akses ke sejumlah besar perpustakaan dan aplikasi yang kompatibel. Dari sekian banyak metode yang telah diaplikasikan untuk deteksi wajah, metode yang dipakai untuk penelitian ini  adalah metode Viola Jones. Dari hasil penelitian ini, diperoleh nilai keberhasilan sebesar 60%. Nilai ini diperoleh setelah melakukan percobaan sebanyak 40 kali, dengan keberhasilan deteksi oleh sistem sebanyak 24 kali. Hasil akhir menunjukkan bahwa intensitas cahaya sangat mempengaruhi peforma dari sistem. Ketika intensitas cahaya bernilai 8 Lux didapatkan tingkat akurasi sebesar 30 %, sedangkan ketika intensitas cahaya bernilai 40 Lux maka didapatkan tingkat akurasi sebesar 90%.

No abstract