The case of motorbike theft is one of the common problems in the community that needs to be found a solution. One of solutions to secure motorbikes from potential theft is to install smart vehicle technology (smart vehicle). This paper describes the design of IoT vehicle safety systems using Arduino Mega microcontroller, fingerprint sensor, ESP8266 and Blynk applications on smartphones. For the experiment, this study uses an automatic transmission motorbike that represents the public motorcycle model. For research methods, this study uses three stages of design. The first stage is to compile the prototype hardware of a motorcycle security system using a fingerprint sensor, and a microcontroller. The second phase, IoT that uses a notification system on ESP8266-based smartphones and Blynk applications are designed. In the third stage, the security system and notification system sent via the smartphone are combined. In the results of the first phase of the trial, five people (whose fingerprints were registered and not registered) were involved to show the system was working as we expected. This means that even if the vehicle ignition is in the "ON" position, the motorcycle engine cannot be started using an electric starter method or using a crank start system. To be able to "start" a motorcycle engine, the user's fingerprint must be verified first. In the second phase of the trial, notification via the Blynk application on the smartphone will show someone's fingerprint that started the motorcycle has been verified or not verified. Finally in the final stage of the trial, information systems in the form of notifications on smartphones can be realized to inform users who are trying to start the motorcycle engine have been verified or not verified. The results of this study are expected to become IoT applied references for motorcycle and other vehicle security systems.

Abstract

Akurasi sebuah klasifikasi citra ditentukan oleh pengklasifikasi.  Meskipun RoI (Region of Interest) tidak menentukan secara langsung akurasi, namun RoI menentukan lingkup klasifikasi citra.   Terdapat tiga algoritma yang dapat digunakan sebagai algoritma RoI yaitu; Balanced Histogram Thresholding (BHT), algoritma Otsu, dan algoritma klasterisasi K-Means.  Paper ini meninjau algoritma Otsu dan algoritma klasterisasi K-Means yang digunakan oleh lima peneliti.  Dari ke lima peneliti; tiga peneliti menerapkan algoritma Otsu dan dua peneliti menerapkan algoritma K-Means sebagai algoritma RoI. Setelah operasi RoI, ke lima peneliti menerapkan algoritma GLCM (Gray Level Co-occurance Matrix) sebagai pengekstraksi ciri tekstur.  Hasil ekstraksi ciri diklasifikasi dengan menggunakan berbagai pengklasifikasi antara lain SVM (Support Vector Machine), Naive Bayes, dan Decision Tree. Akhirnya dengan membandingkan hasil dari ke lima peneliti, akurasi tertinggi diperoleh sebesar 100% dengan pengklasifikasi SVM menggunakan algoritma Otsu sebagai algoritma RoI, dan akurasi terendah adalah sebesar52% yang menggunakan algoritma Otsu pada kanal S dari citra HSV (Hue, Saturation Value).

Abstract-Odor sensing technology in robotic research introduce two research field namely odor recognition and odor source localization. Odor source localization research also includes the odor recognition ability with localization method. This paper shows some experiment had been done to localize odor source using single agent and multiple agents. Experiment shows that single agent can't be used in dynamic environment, hence also can't be used in real life application. This paper promotes an algorithm known as Particle Swarm Optimization (PSO) to solve these problems. The experiment