Determining Opacity By Machine Vision.

Opacity is recognized as being one of the most important aspects of the appearance of an object. therefore, assessment of the opacity is a common process of quality control in various industries. the simplest and most widely used method of measuring the opacity of a surface is calculating the contrast ratio (cr) criteria; however, the traditional process of the calculation of the cr requires a spectrophotometer or a colorimeter, which is costly and time-consuming approach. so, finding an efficient and a simple alternative method to measure the opacity has received much attention in the past decades. in the present research, a machine vision technique was used to achieve this goal. in the first place, a picture of 35 translucent samples was provided using a digital scanner. the measurements were carried out on both a black and a white background. the cr values of the samples were calculated using various terms representing the lightness in different color spaces. in each case two different calculations were done based on the mean and the mode values of the data. the results confirmed the consistency of the proposed method with the traditional spectroscopic method with an error of 1.3 percent at cr values between 0.9 to 1

Determining opacity by machine vision.

Opacity is recognized as being one of the most important aspects of the appearance of an object. Therefore, assessment of the opacity is a common process of quality control in various industries. The simplest and most widely used method of measuring the opacity of a surface is calculating the contrast ratio (CR) criteria; however, the traditional process of the calculation of the CR requires a spectrophotometer or a colorimeter, which is costly and time-consuming approach. So, finding an efficient and a simple alternative method to measure the opacity has received much attention in the past decades. In the present research, a machine vision technique was used to achieve this goal. In the first place, a picture of 35 translucent samples was provided using a digital scanner. The measurements were carried out on both a black and a white background. The CR values of the samples were calculated using various terms representing the lightness in different color spaces. In each case two different calculations were done based on the mean and the mode values of the data. The results confirmed the consistency of the proposed method with the traditional spectroscopic method with an error of 1.3 percent at CR values between 0.9 to 1