Defining Problem: Segmentation of Fruit by Extracting Natural Images using Image Processing
issue 1

Defining Problem: Segmentation of Fruit by Extracting Natural Images using Image Processing

Amit Welekar 1, Apurva Pantawane2, Nikita Vairagade3, Rutuja Panchgade4, Harsha Raut5, AkshataKumbhare6,  

 1, 2, 3,4,5,6, Department of computer Engineering, Bapurao Deshmukh College of Engineering, Sevagram, Wardha

Abstract –

This paper into define the problem related to segmentation of fruit images captured in natural environment. The image captured in natural environment is contaminating due to the unwanted lighted intensity. So we are using to different algorithm for different kind of color image. The define problem and discuses in this paper.

Index TermsImage segmentation, clustering.


Image segmentation plays important role in computer vision application particularly for fruit grading system. In general, segmentation technique emulates the abilities of humans in recognizing objects. In addition, this technique offers non-destructive method for classifying objects and produces more consistent result than humans. image segmentation is that the process of partitioning a digital image into multiple segments . This process is required to make sure that the interest area could be perfectly and correctly isolated from the background area with the aim to increase the accuracy in object identification phase. An incorrect segmentation degrades the segmentation process and thus will produce poor end in object analysis. There are several segmentation techniques that can Image segmentation is an important phase in image processing where it is used to separate a digital image into several areas. Of all the techniques, thresholding and clustering are extensively used in agricultural area. However, it has shown that these conventional techniques were inadequate for segmenting fruit images captured in natural environment. The main purpose of this process is to remove the background in order to ensure that only the area of interest is processed during the object analysis phase. An accurate segmentation result is very significant for the success in image analysis. Image segmentation which was widely used in several disciplines such as agricultural and medical can be performed by various techniques. There are two segmentation techniques that were extensively used by many researchers. The techniques are Improve thresholding and Adaptive K- means clustering. The admiration of thresholding technique among the researchers is because of its simplicity and straightforwardness. This technique is simple because it separates the digital image into numerous areas based on the gray levels of the image. The original images which were converted into the Gary scale format can be easily operated to produce more consistent segmented images. Technology advancement on the image segmentation technique has experienced tremendous growth both in theory and application.


The machine vision in image segmentation using traditional segmentation techniques and proposed an improved technique for segmenting images captured under natural environment.  Image segmentation refers to a process of partitioning a digital image into multiple regions with the aim to extracts object of interest from the background. However, the segmentation process is very challenging especially for experiment which conducted in outdoor environment. Sharifah Lailee Syed Abdullaha, Hamirul’Aini Hambalia,” Segmentation of Natural Images Using an Improve Thresholding-based Technique”, 41            (2012) 938 – 944.[1]

Image processing is effective tool for analysis in various fields and applications in agriculture. Today’s very advanced and automated industries used more accurate method for different inspection processes of agriculture object. This task known as robotics Indian  agriculture industry many kind of activities are done like quality inspection, sorting, assembly, painting, packaging. Above mentioned activities are done manually by using Digital Image processing tasks done conveniently and easily. Using Digital image processing many kind of task fulfils like object Shape, size, Color detection, texture extraction, firmness of object, aroma, maturity Hue Zhang* Jason E. Fruits, Sally A Goldman. “A Fast Texture Feature Extraction Method for Region-based Image Segmentation. [2]

Segmentation is the first step in analyzing or interpreting an image automatically. In particular applications, like image compression or image recognition, entire image can’t be processed directly. Hence many segmentation techniques are proposed to segment an image before processing it..  They are either applied for grading or inspecting quality of food products and Fruits. These developed techniques use thresholding and clustering approach to get proper segmented output. In this paper an image segmentation approach is developed based on k-means adaptive clustering. Haniza, Y.Hamzah, A., & Hazlita, M. I., “K-mean Clustering for Segmentation of Irregular Shape Fruit Images under Various Illumination,” Measurement, vol. 45, 15991608, 2012.[3]

Image segmentation is the fundamental approach of digital image processing. Among all the segmentation methods, Otsu method is one of the most successful methods for image thresholding because of its simple calculation. Otsu is an automatic threshold selection region based segmentation method. This paper studies various Otsu algorithm .Krishnaveni, M., & Radha, V., “A Review on Otsu Image Segmentation Algorithm” International Journal of Engineering Science and Technology (IJEST), vol. 3(2), pp. 1014-1020, 2011. [4]

Segmentation refers to a technique in which an image in digital form is partitioned into multiple segments (basically groups of pixels, also termed as Super pixels). This paper is a survey on Image Segmentation with its clustering techniques. Image Segmentation is the procedure of apportioning a picture into numerous segments, to change the exemplification of a picture into another which is more useful and easy to segment. A few universally useful calculations and approaches have been developed for picture division. It separates a digital picture into numerous locales to investigate them. It is likewise used to recognize segment items in the picture. Priyansh Sharma1 and Jenkin Suji,” A Review on Image Segmentation with its Clustering Techniques,” International Journal of Signal Processing, Image Processing and Pattern Recognition Vol.9, No.5 (2016), pp.209-218.[5]

As the premise of feature extraction and pattern recognition, image segmentation is one of the fundamental approaches of digital image processing. This paper enumerates and reviews main image segmentation algorithms, then presents basic evaluation methods for them, finally discusses the prospect of image segmentation. Some valuable characteristics of image segmentation come out after a large number of comparative experiments. Wen-Xiong Kang, Qing-Qiang Yang, Run-Peng Liang,” The Comparative Research on Image Segmentation Algoriths,” First International Workshop on Education Technology and Computer Science 2009 [6]


Image segmentation is performed to separate out the object of interest and background. Image attainment is an important task in image segmentation, according to “One-line fruits grading according to their external quality using machine vision. In both these approaches initial cluster value is required, but both are sensitive to find the initial cluster. Is able to perform segmentation of circular shaped fruit images captured under natural illumination but the same approach cannot be used for irregular shaped fruit images.

 Image Segmentation is a fundamental approach of this problem Statement of fruit images capture in natural light. Previous problem the improve threshold Algorithm works as only circular shape and they used only the two colors (Yellow, Green). The Proposed System includes three phases namely; Image Acquisition, Pre-Processing, Image Segmentation of the fruits.


Image Segmentation is a fundamental approach of Digital Image Processing. Image Segmentation may be a process of dividing a picture into distinct regions with the aim to extracts object of interest from the background. This process may be a crucial stage within the image analysis where the end in this stage influences the performance of the whole process. Segmenting the chosen region ensured that only the thing of interest was processed during the color analysis phase. A false segmentation will cause degradation of the object measurement and classification process.

In this Project, Segmentation is the first step in analyzing or interpreting an image automatically. In particular applications, like image compression or image recognition, entire image can’t be processed directly. Hence many segmentation techniques are proposed to segment an image before processing it. This made it possible to develop many techniques which are currently using in different industries and agriculture field.

Figure .1 Architecture of segmentation of fruit by extracting natural image using image processing.

Image Acquisition:-

An image is analyzed as it is clicked. The user is given tools to discard the consider noises. The image acquisition is completed employing a camera and it’s loaded and saved using MIL software. MIL works with images captured from any type of color (RGB) or monochrome source (Gray). MIL supports the saving and loading of images. It supports file formats like TIF (TIFF), JPG (JPEG), BMP (bitmap), also as raw format. The input image got as a RGB image.

 Pre-processing: –

The pictures  are obtained during image acquisition might not be directly suitable for identification and classification purposes due to some factors, like noise, weather , and poor resolution of images and unwanted background. We tried to adopt the established techniques and study their performances.

The steps involved in per-processing are

A. Input image

B. Background subtraction

C. Converting RGB to gray

D. Converting gray to binary

E. Filtering

Image Segmentation:-

Image segmentation refers to a process of partitioning a digital image into multiple segments or regions to simplify the representation of a picture. Segmenting the selected region ensured that only the object of interest was processed during the color analysis phase.

 RGB Image:-

RGB is one among the formats of color images. Here the input image is represented by three matrices of sizes regarding the image format. The three matrices in each image correspond to the colures red, green and blue and also say that of how much of each of these colors a certain pixel should use.

Image Pre-processing in this phase, all the RGB images were resized into 300×300 pixel resolutions. This process is required to decrease the processing time and to ensure that all the images have the same pixel intensity value. Image segmentations refer to a process of partitioning a digital image into multiple segments or region to simplify the representation of an image. Segmentation the selected region ensured that only the object of interest was processed the color analysis phase.

Background Subtraction:-

Background subtraction is a process of extracting foreground objects in a particular scene of an image. A foreground object is defined as an object of attention which helps in reducing the amount of data to be processed.


Fifty fruit images have been collected for fruit segmentation system. These fruit images have been divided into training fruit set testing fruit set, where 36of the collection fruit images were to develop and train the system. The training fruit images are required be sent in and processed by system when develop the segmentation algorithm for the system. The mean color values of the fruit are often computed after the user crop the world of fruit within the fruit image. The system will compute the mean values for every of the red, green and blue (RGB) component of the cropped fruit area by manipulating and computing on the 3D matrices that stored all of the fruit pixels.

 Analysis Methods:-

The fruits recognition system consists of 5 main processing modules, which are, fruit input selection module, fruit color computing module, fruit shape computing module, fruit size computing module, and fruit classification or recognition module. The first processing module of the system will prompt the user to pick a fruit image from the fruit selection menu for further recognition process. The fruit color-computing module is important to perform the fruit color feature extraction tasks. Subsequently, the fruit shape-computing module will analyze the fruit then fruit region feature properties.


In this paper we concluded that image segmentation is a process that divides a digital image into separate regions with the aim to obtain only the interest object and remove the background. There are several segmentation techniques which are utilized in object identification like thresholding and clustering techniques. Therefore, this researcher has produced an innovative segmentation algorithm for fruit images which is in a position to extend the segmentation accuracy. The algorithm is an integration of modified thresholding and adaptive K-means method. The integration of both methods is required to increase the segmentation accuracy for fruit images with different surface color. The result showed that the innovative method is able to segment the fruit images with high accuracy value.


  1. X. P. B. Artizzu, A. Ribeiro, A. Tellaeche, G. Pajares, and C. F. Quintanilla, “Analysis of natural images processing for the extraction of agricultural elements,” Image and Vision Computing, vol. 28, pp. 138-149, 2010.
  2. Seng, Woo Chaw, and Seyed Hadi Mirisaee.”A new method for fruit recognition system. “Electrical Engg and Informatics 2009.ICEEI’09. International conference on Vol.1.IEEE, 2009.
  3. Susovan Jana, Saikat Basak, “Automatic Fruit Recognition from Natural Images using Color and Texture Features “.Research Gate 2017.
  4. Acuña, R. G. Gonzalez, Junli Tao, and Reinhard Klette. “Generalization of Otsu’s binarization into recursive colour image segmentation.” In Image and Vision Computing New Zealand (IVCNZ), 2015 International Conference on, IEEE, 2015 pp.  1-6.
  5.  L. H. Thai, T. S. Hai, Nguyen Thanh Thuy. “Image Classification using Support Vector Machine and Artificial Neural Network” International Journal on Information Technology and Computer Science, 2012, 5, 32-38.
  6. Mr.A. R. Welekar, Ms Monali Dahapute (ICMTEST 2016) “K-means Clustering for Segmentation of Irregular Shape Fruit Images under Various Illumination.”
  7. H. M. Moftah, A. TaherAzar, E.T. Al-Shammari, N. I. Ghali, A. E. Hassanien and M. Shoman, ‖ Adaptive k-means clustering algorithm for MR breast image segmentation‖, Neural Comput & Applic vol. 24, (2014), pp.1917–1928.
  8.  M. Xess and S. A. Agnes, Survey On Clustering Based Color Image Segmentation And Novel Approaches To Fcm Algorithm, IJRET: International Journal of Research in Engineering and Technology, pp. 346-349.
  9.  Akanksha Bali and Dr. Shailendra Narayan Singh,” A Review on the Strategies and Techniques of image Segmentation”ASET, Amity University, Noida, India,pp 2015.
  10.  Susmita Sahu, Himadri Sarma, Dibya Jyoti Bora. Image Segmentation and Its Different Techniques: An In-depth Analysis” MCA Student, School of Computing Sciences, Kaziranga University Jorhat, India, pp. 2018.
  11. Miss Hetal J. Vala, Prof. Astha Baxi, “A Review on Otsu Image Segmentation Algorithm” Parul Institute of Engineering & Technology, Waghodia, Gujarat (India).ISSN:2278-1323, 2013.
  12.  Kass M, Witkin A, Terzopoulos D. Snakes: Active contour models. International Journal of Computer Vision. 1988; 1:321-331. DOI: 10.1007/BF00133570.
  13.  Martin David Levine,” Range Image Segmentation Based on Differential Geometry: A Hybrid Approach” IEEE Transactions on Pattern Analysis and Machine Intelligence · July 1989.

Related posts

Performance analysis and implementation of spectrum sensing of interleaved signals in SC-FDMA using Cognitive radio


Contrast & Color Improvement Based Haze Removal Of A Underwater Image Using Fusion Technique


Handwritten Character Recognition Using Deep Learning


Leave a Comment