Video surveillance systems obtain a great interest as application-oriented studies that have been growing rapidly in the past decade. The most recent studies attempt to integrate computer vision, image processing, and artificial intelligence capabilities into video surveillance applications. Although there are so many achievements in the acquisition of datasets, methods, and frameworks published, there are not many papers that can provide a comprehensive picture of the current state of video surveillance system research. This paper provides a comprehensive and systematic review on the literature from various video surveillance system studies published from 2010 through 2019. Within a selected study extraction process, 220 journal-based publications were identified and analyzed to illustrate the research trends, datasets, methods, and frameworks used in the field of video surveillance, to provide an in-depth explanation about research trends that many topics raised by researchers as a focus in their researches, to provide references on public datasets that are often used by researchers as a comparison and a means of developing a test method, and to give accounts on the improvement and integration of network infrastructure design to meet the demand for multimedia data. In the end of this paper, several opportunities and challenges related to researches in the video surveillance system are mentioned.

Confusion and diffusion are two main concepts in cryptography that affect the quality of encryption and resistance to various attacks. Both of these concepts are influenced by the quality of the keystream used. Keystream generation needs to involve plain images as one way to bind sensitivity to plain images. But when it involves plain images directly, encryption security has negative side effects. So this research proposes to perform a diffusion operation with a logistic map on a plain image before generating the keystream. Not only using one keystream, the proposed method also uses three other keystreams and Arnold transform whose input parameters are processed by hash operations. Two keystreams are generated using the four-dimensional Lorenz system for the permutation of column and row image pixels. Next, sixteen Arnold transforms with different parameters are used for bit permutation. The last keystream is generated with a Henon map to carry out the diffusion stage two so that a cipher image is produced. Based on the testing results, the proposed method has satisfactory encryption results that are resistant to various attacks, especially brute force, statistical, and differential attacks.

The importance of early fire detection can help in providing warnings and avoid disaster that led to the economic damage and loss of life. Fire detection techniques with conventional sensors have limitations, which require a long time to detect a fire, especially in a large room and cannot work in the open area. This study proposed a fire detection method, to detect the possibility of fire based on visual sensor using multi-color feature such as color, saturation, luminance, background subtraction and time frame selection for fire detection. The evaluation in this studies conducted by calculating the error rate of the fire detection.

Coconut tree grows rapidly in tropical region such as Indonesia. Coconut wood is used as alternative or complementary raw material for housing or making furniture. Abundant coconut trees are planted, however the utilization of coconut wood as raw material for furniture is still very rare in Indonesia. This is caused by the low quality of coconut wood, since it has not found adequate technology for the processing of coconut wood. This paper presents our experimental work on coconut wood quality classification using self-tuning MLP classifier (AutoMLP) and Support Vector Machine (SVM). For SVM classifier we used the LibSVM library, available in RapidMiner. The Gray-Level Co-occurrence Matrix (GLCM) is used to extract the texture features of coconut wood images. Experiment result shows that AutoMLP gives the best accuracy rate at 78.82%, which is slightly better than 77.06% of SVM.