At this time, many illegal activities are being been carried out, such as illegal mining, hunting, logging, and forest burning. These things can have a substantial negative impact on the environment. These illegal activities are increasingly rampant because of the limited number of officers and the high cost required to monitor them. One possible solution is to create a surveillance system that utilizes artificial intelligence to monitor the area. Unmanned aerial vehicles (UAV) and NVIDIA Jetson modules (general-purpose GPUs) can be inexpensive and efficient because they use few resources. The problem from the object-detection field utilizing the drone’s perspective is that the objects are relatively small compared to the observation space, and there are also illumination and environmental challenges. In this study, we will demonstrate the use of the state-of-the-art object-detection method you only look once (YOLO) v5 using a dataset of visual images taken from a UAV (RGB-image), along with thermal infrared information (TIR), to find poachers. There are seven scenario training methods that we have employed in this research with RGB and thermal infrared data to find the best model that we will deploy on the Jetson Nano module later. The experimental result shows that a new model with pre-trained model transfer learning from the MS COCO dataset can improve YOLOv5 to detect the human–object in the RGBT image dataset.

The high dimensionality in big data need a heavy computation when the analysis needed. This research proposed a dimensionality reduction using deep belief network (DBN). We used hyperspectral images as case study. The hyperspectral image is a high dimensional image. Some researched have been proposed to reduce hyperspectral image dimension such as using LDA and PCA in spectral-spatial hyperspectral image classification. This paper proposed a dimensionality reduction using deep belief network (DBN) for hyperspectral image classification. In proposed framework, we use two DBNs. First DBN used to reduce the dimension of spectral bands and the second DBN used to extract spectral-spatial feature and as classifier. We used Indian Pines data set that consist of 16 classes and we compared DBN and PCA performance. The result indicates that by using DBN as dimensionality reduction method performed better than PCA in hyperspectral image classification.

Internet technology led to the emergence of Web 2.0 which increase the number of User Generated Content (UGC) on the network. Online product review is a form of UGC. The case study in this research is a review of handphone products. The large number of reviews will take long time to read and compare between existing product reviews, so we need a technique that online product review can be read quickly without losing of its important information. The technique that can be used is the text summarizing technique. Text summarization techniques produce simplified versions of texts. In general, text summarization can be divided into two types, namely extractive and abstractive summaries. This research used extractive summaries. One important component in the process of obtaining an extractive summary is sentence extraction. In this study, the algorithm used for sentence extraction is TextRank. The purpose of this study was to determine the performance of the TextRank algorithm with handphone product reviews data by implementing it in different data conditions based on the presence or absence of a stopword and typo. These data conditions are used to formulate test scenarios. Testing is done by calculating the Rouge-1 value which compares the summary of system and experts. Expert who involved in this study are 2. Expert 1 is a person with expertise in Indonesian and Expert 2 is someone who has the knowledge and understanding of mobile phones with various types and characteristics. From the test results obtained, Expert 1 gets the best results for scenario 2 where data conditions are there is typo and no stopword with an average value of Rouge-1 of 42.29% and Expert 2 gets the best results for scenario 3 where data conditions are no typo and there is stopword with an average value of Rouge-1 is 46.71%. The results shows that the TextRank algorithm is not able to produce a good summary for handphone product review dataset.

Smart Farming (SF) has garnered interest from computer science researchers for its potential to address challenges in Smart Farming and Precision Agriculture (PA). This systematic review explores the application of Fuzzy Logic (FL) in these areas. Using a specific anonymous search method across five scientific web indexing databases, we identified relevant scholarly articles published from 2017 to 2024, assessed through the PRISMA methodology. Out of 830 selected papers, the review revealed four gaps in using FL to manage imprecise data in Smart Farming. This review provides valuable insights into FL for potential applications and areas needing further investigation in SF.

Segmentation of fetal organs such as head, femur, and humérus on ultrasound image is one of the challenges in realization of automated system for fetal biometry measurements. Although many methods have been developed to overcome this problem, most of them are generally specific to one organ of the body alone. The research in this paper will focus on a machine learning method that has been available before: multilayer super pixel classification using random forest. The focus of this study is to improve the accuracy by exploring compactness parameter in the formation of super-pixels. In addition, we also add moment image features such as translation, rotation, and scale invariant to improve the segmentation performance. The experimental results showed that the difference in compactness parameters will provide different result for the accuracy, Fl-score, recall, and specificity. The addition of moment features can also improve the performance of image segmentation of fetal organs even though increase was not significant. Fetal head segmentation using proposed method has higher Fl-score and specificity, but lower accuracy and recall compared to previous methods. Whereas fetal femur and humérus segmentation using proposed method has higher accuracy, Fl-score, recall and specificity compared to previous method.