To examine how particle loss affects the evaluation of masticatory ability using the image analysis method. The subjects were divided into two groups:Y-group (healthy young adults: 10 male, 10 female; mean age 27.6 ± 1.9 years), and D-group (denture wearing older adults: 13 male, 9 female; mean age, 75.1 ± 5.3 years). Raw carrots, peanuts, and a combined test food from 5 daily food materials (mixed foods) were selected as test foods. Images of the boluses were captured and processed after the subjects freely masticated the test foods until the point of swallowing or completed half strokes of total chewing cycles. The median particle size (X50) was calculated from the data for each particle. We compared X50 calculated from all particles obtained from the masticatory bolus (as X50N) with the X50 of only selectively larger particles (as X50R). Significant correlations were observed between X50N and X50R (p < 0.01) determined in almost all masticatory conditions. In particular, for particle sizes of 2 mm or more, the correlation coefficient between X50N and X50R was notably strong (r > 0.75). The method of analyzing only larger particles makes it possible to evaluate masticatory ability without losing the characteristics of the original particle size distribution of the entire bolus. This finding can be applied for the evaluation of masticatory ability, especially among older adults who have difficulty retrieving the total amount of test food due to decreased activities of daily living.