In this paper we present a comprehensive review of the state of the art of smoke and fire detection techniques using image processing. Smoke is a good indicator of a pre-fire condition and many fires are indicators of subsequent dangerous situations due to the spread of fire. In this paper we first start our comparison of smoke detection methods and different types of approaches for the classification of smoke. Furthermore we analyse different types of technologies and various models involve in detection techniques such as RGB and HSI models for detecting smoke and fire. Generally, the false alarm rate can be reduced by image processing through effective types of detection techniques such as vision-based or sensor-based methods. Mainly in this paper, we focus on optimised technologies in order to detect smoke and fire at the earliest possible stage of the event, and smoke and fire detection by satellite vision methods.

This work is to explicate the data collected during the turning of AISI 1045 alloy steel components in near dry condition with emulsified cutting fluids prepared from cooking oils such as Palm oil and Peanut oil. The base oils are tested for its relative density, viscosity and flash point following ASTM standards . Highly influencing turning factors are identified and the experiments are planned and arranged using Taguchi's L 27 (3 5 ) orthogonal array, the experiments are repeated to reduce the errors. The quality aspect of machined components and the machining interface temperature is observed as the outcomes. The prediction models are created for the experiments through regression analysis.

No abstract

In this paper we present a comprehensive review of the state of the art of smoke and fire detection techniques using image processing. Smoke is a good indicator of a pre-fire condition and many fires are indicators of subsequent dangerous situations due to the spread of fire. In this paper we first start our comparison of smoke detection methods and different types of approaches for the classification of smoke. Furthermore we analyse different types of technologies and various models involve in detection techniques such as RGB and HSI models for detecting smoke and fire. Generally, the false alarm rate can be reduced by image processing through effective types of detection techniques such as vision-based or sensor-based methods. Mainly in this paper, we focus on optimised technologies in order to detect smoke and fire at the earliest possible stage of the event, and smoke and fire detection by satellite vision methods.

No abstract