Machine vision based sorting of Chikoo (sapota) fruit of Dahanu taluka region using Digital Image Processing

ABSTRACT:

In this paper we propose a system with a general approach is developed to estimate the maturity (ripeness) level of Chikoo (Sapota) fruit while not touching it. Sapota fruit maturity level usually makes farmers confused once selecting a sapota that encompasses a sensible maturity. Sometimes, even farmers still use manual strategies to differentiate level of maturity, whereas the approach that human labor is commonly inaccurate and totally different in its determination. The distinction is because of the various perceptions of every person. From these issues then the necessity of machine sorting system on agriculture is felt necessary. Therefore, researchers can conduct research on sortation system. The fruits are often classified consistent with completely different conditions as pre-harvest, postharvest harvest, storage conditions in controlled setting or harsh setting. during this study the fruit beneath take a look at were classified at the time of harvest in four completely different categories as inexperienced, ripe, ripe and spoiled employing a net camera based mostly laptop vision system. This paper presents a straightforward technique that uses a mix of digital net camera, laptop and indigenously developed graphics software system to live and analyze the surface color of fruits for ripening state recognition. The popularity was done by the closest neighbor classifier engine uses the HSI color distribution in chosen ROI of fruits. Experimental results on a dataset of Chikoo (Sapota) fruits from four completely different ripening states make sure the effectiveness of the projected approach.

PROJECT OUTPUT VIDEO:

PROPOSED SYSTEM:

• In our proposed work, sorting types of sapota fruit images are taken into consideration.

• Initially, the fruit image specified region is cropped and then image is preprocessed using the wiener filter and median filter. Fruit image motion blur is removed by using wiener filter and noise is removed by using median filter.

• Then filtered image is segmented by applying HSI color model.

• The next process is to extract the features such as the statistical color features such as mean and standard deviation for providing more accuracy than the existing systems.

• At last, K-Nearest Neighbour based classification is employed for sorting of sapota fruits.

ADVANTAGES OF PROPOSED SYSTEM:

• The proposed intelligent system improves accuracy rate.
• Robustness
• Efficiency

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System            : Pentium Dual Core.

• Hard Disk       : 120 GB.

• Monitor           : 15’’ LED

• Input Devices  : Keyboard, Mouse

• Ram                 : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : MATLAB.

• Tool                       : MATLAB R2013A