Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

This paper proposes an automatic tomato sorting algorithm using image processing technique in MATLAB. Images of the tomatoes were acquired using a digital camera in a controlled environment. The system made use of histogram equalization, median filter, edge detection, RGB colour and the Grayscale image to acquire the physical parameters of the processed tomato image such as major axis length, centroid and the diameter used for sorting the tomatoes into small, medium and large sizes. Ninety six images of the tomatoes were used to train the algorithm and subsequently thirty different sizes of tomatoes were captured and processed to evaluate the performance of the algorithm, and the result was validated using similar metrics gotten from physical measurements using a Vernier Caliper. The algorithm classified small, medium and large size tomatoes as 10, 9 and 11 numbers respectively; while visual classification based on physical measurements classified small, medium and large size tomatoes as 10, 10 and 10 numbers respectively. This result showed that the system could successfully extract the physical parameters from the images of the tomatoes which were used in sorting them. The process time for each image was 20 seconds.