International Journal of Computer Applications (0975 – 8887) Volume 130 – No.16, November2015 6 Color, Size and Shape Feature Extraction Techniques for Fruits: A Technical Review Amish Patel Assistant Professor Uka Tarsadia University, Bardoli, Gujarat Puja Kadam Assistant Professor Uka Tarsadia University Bardoli, Gujarat Sapan Naik Assistant Professor Uka Tarsadia University Bardoli, Gujarat ABSTRACT Grading of agricultural produce especially the fruits and vegetables has become a perquisite of trading across borders. In India mostly fruit and vegetable growers grade the fruit manually. Manual grading was carried out by trained operators who considered a number of grading factors and fruit were separated according to their physical quality . Manually grading was costly and grading operation was affected due to shortage of labor in peak seasons. Human operations may be inconsistent, less efficient and time consuming. New trends in marketing as specified by World Trade Organization (WTO) demand high quality graded products. Farmers are looking forward to having an appropriate agricultural produce-grading machine in order to alleviate the labor shortage, save time and improve graded product‟s quality.. The need to be responsive to market demand places a greater emphasis on quality assessment, resulting in the greater need for improved and more accurate grading and sorting practices. Size variation in vegetables like potatoes, onions provided a base for grading them in different categories. Every vegetable producing country had made their own standards of different grades keeping in view the market requirements. Keywords Fruit grading, Color Feature Extraction, Shape Feature Extraction, Size Feature Extraction. 1. INTRODUCTION India is an agricultural country, where about 70% of people spend their lives by agriculture. Farmers have wide range of fruits and vegetable crops to harvest. The quality and optimal production of crops highly depends upon scientific reasons [16]. The management of permanent fruit crops requires close monitoring especially for the managing of diseases that can affect fabrication considerably and consequently the post harvest life. The image processing can be highly applied on agricultural applications for various purposes like: 1. To detect diseased leaf, stem, fruit and roots. 2. To determine size & shape of fruits and plant. 3. To estimate chlorophyll content of a plant. 4. Measurement of plant leaf area. 5. Weed detection The fruits sorting and grading are considered the most important steps of handling. Sorting and grading are major processing tasks associated with the production of fresh- market fruit, vegetable and crop types. Post harvest process of fruits and vegetables is considered as the most important process that leads to conserve the quality until reach to the consumers. Sorting is a separation based on a single measurable property of raw material units, while grading is “the assessment of the overall quality of a food using a number of attributes”. Sorting of agricultural products is accomplished based on appearance (color and absence defects), texture, shape and sizes [17]. Manual sorting is based on conventional visual quality inspection performed by human operators, which is tedious, time-consuming, slow and non-consistent. It has become increasingly difficult to hire personnel who are adequately trained and willing to undertake the tedious task of inspection. A cost effective, consistent, superior speed and accurate sorting can be achieved with machine vision assisted sorting. Grading of fruits, vegetables and crops is a very important operation as it fetches high price to the grower and improves packaging, handling and brings an overall improvement in marketing system. They are generally graded on the basis of size and graded products are more welcome in export market. Grading could reduce handling losses during transportation. Grading based on size consists of divergent roller type principle having inclination, expanding pitch type, inclined vibrating plate and counter rotating roller having inclination type graders. Weight grading based on density and specific gravity of agricultural commodities In this paper, we have given brief overview on how fruits and vegetables can be graded and sorted. In the second section shape feature extraction methods are discussed, in third section color feature extraction techniques are described, following size extraction methods in fourth section are conversed. In fifth section conclusion has been given after the review of different techniques based on size, shape and color feature extraction using image processing. 2. SHAPE FEATURE EXTRACTION AND ALGORITHMS In [1] Hatou used a triangulation laser range finder for constructing the 3-D shape of tomato fruits as they traveled on a conveyor belt beneath the range finder or scanner. Initially, a reference or ideal tomato shape was built from measurements obtained using the same laser range scanner. The inspected shape of each fruit was compared with the reference shape, and the differences between the two shapes were used for classification. Using an intelligent classifier system which included neural networks and expert system this was performed. The authors originated that the grading results of their system were similar to those achieved by a skilled human inspector. The disadvantage was that the method was time-consuming – it took 5 s to classify each tomato. Passive 3-D machine vision methods are recognized in the literature as shape or range from X techniques, where X represents different 3-D cues such as stereo, shading, silhouettes or occluding contours, motion, contour or shape, shadows or darkness, texture, and fractal geometry. From one or more 2D images these 3D cues can be determined. [2] It is sufficient to say that surface orientation can be obtained from the range by taking