APPLICATION OF COMPUTER VISION FOR COUNTING OIL PARTICLE CONTAMINANT

Abstract

Contaminant particles in the oil are an indication of the condition of a part or engine. The large number of particles in the oil provides information that it is necessary to check and maintain the engine. If the particles in the oil are left unattended, there will be damage to parts, even worse, engine failure. Therefore, it is important to know how many particles are in the oil and ISO provides an assessment criteria for the cleanliness of the oil, which is determined based on the particle size and the number of particles. In this study, particles whose number and size were calculated were obtained from a membrane that had been poured with 100 ml of oil, then under a microscope, the image of the membrane was captured and using computer vision to determine the particle size and number of particles, after that the expected output was the criteria based on ISO Cleanliness 4406 table. Based on the experimental results, the computer vision application can detect the number and size of particles in the image and provide results in accordance with the ISO Cleanliness table. Computer vision applications can be used to calculate the number of particles in the oil.