DETECTION OF STONE WEEVIL IN MANGO FRUIT USING NON-DESTRUCTIVE TECHNIQUE

Transcription

1 DETECTION OF STONE WEEVIL IN MANGO FRUIT USING NON-DESTRUCTIVE TECHNIQUE SPOORTI SAMBRANI, SOURABH DHUMWAD, BHAGYASHRI NAVALLI, SHIVRAJ H J, SUVARNA G KANAKARADDI Department of Computer Science and Engineering,BVBCET Hubli,India ABSTRACT The king of fruits Mango is very nutritious and rich in carotenes. India produces 50 percentage of the total worlds mango. Many researchers have reported the maturity indices and quality parameters for determination of harvesting time and eating quality. The methods currently used for determination of quality of mango are mostly based on biochemical analysis, which leads to destruction of the fruits. The quality attributes of alphonso include freedom from external damages such as latex or sap injury and decay, bruises, uniform weight, color, aroma, firmness, shape and size. The mango seed weevil, Sternochetus mangiferae (Fabricius), is not found Florida, but its presence in the major mango producing areas of the world indicates that it is a potential pest here. To retain the overall mango fruit quality and to reduce postharvest losses during supply chain can be achieved by adopting suitable postharvest novel technologies involving image processing and pattern recognition techniques. To achieve this task the researchers in this area needs the database that contains healthy as well as Spongy Tissue affected alphonso mangoes. In this research an attempt is made to develop a database and percentage of stone weevil in mango fruit is identified using non-destructive approach and Digital X-ray imaging technique. Keywords Digital X-Ray imaging. Non-Destructive Approach, Stone Weevil, Spongy Tissue. I. INTRODUCTION India is an agricultural country. From last thousands of years back the India was identified by the Ayurveda Country and also agriculture based country.its economy is mainly dependent on agriculture and so agriculture is the backbone of India.Mango is a widely exported fruit from India.Mango is commonly eaten as a fresh fruit. The fruit may also be processed to produce mango pickles or chutneys. A high quality variety, Alphonoso suffers from physiological disorder of spongy tissue growing in the pulp portion due to which, its export value and consumer satisfaction is affected. Also certain varieties of mango like Neelam and Totapuri have seed weevil or stone-weevil infestation, which are not visible externally in Indian mangoes. The texture is soft, spongy or leathery depending on the severity of damage. The tissue affected is visible only when the ripe fruit is cut into two halves. This necessitates the investigation defect detection in mangoes. Three strategies are used to control weevil infestation; quarantine, hygiene and chemical control. This paper is an effective and accurate solution that can detect spongy tissue and seed weevil infected mangoes by instrumental means and grading good quality mangoes for consumption and export. This is a totally non-destructive indigenous technology was developed by employing Digital X-ray imaging. Preprocessing of image has been carried out efficiently The digital x-ray scanned mangoes are safe for consumption and there is no health hazard. The mango seed weevil Sternochetus mangiferae (Fabricius), is not found Florida, but its presence in the major mango producing areas of the world indicates that it is a potential pest here. It is spread mainly by transportation of infested fruits since the weevil develops within the mango seed and can therefore be transported unnoticed from one locality to another.organization of the paper is as follows: Section II discusses Defects occurring in mangoes. Section III discusses Literature survey. Section IV explains Proposed work. Section V explains Methodology. Section VI shows the Result and Section VII concludes the paper. II. DEFECTS OCCURRING IN MANGOES 204

2 There are many pests, diseases and disorders occurring in mangoes that damage fruit. But there are few names which are common and which are important to be studied because these causes most harm to the fruit and indirectly it is a loss for the growers and exporters. The detailed information of stone weevil is as follows. A. Mango stone weevil The Mango stone weevil is a major quarantine pest. The eggs are deposited into shallow depressions on young fruit and covered by hardened sap in a crescent shape.freshly laid eggs are creamy-white. They are elliptical, mm ( mm) long and ( mm) wide.the adults are dark brown to grey-black oval shaped weevil. It shows mottled markings on wing covers. The weevil has a prominent snout. Adults are about 10 mm in length. The life cycle is of one generation per year, similar to other native weevils. The Damage to fruit is caused by eggs which are laid into fruit when they are 30 mm in length or larger. Egg laying sites are readily seen on fruit in association with leaking sap. Grubs move directly into developing seed and do not damage the flesh. Adults may occasionally be seen moving onto branches and the outer canopy at flowering. Take random samples of fully developed fruit, and cut fruit to inspect the seed for larvae or pupae (see Fig. 1). Adults emerge from fruit within 2 months of fruit fall. The Control Strategies include removing fallen fruit reduces populations in infested areas. Hence exporters and processors are unable to provide quality mangoes and pulp free from the internal disorders. As a result, many countries do not import mangoes. (a) (b) Figure1: Pests of Mango (a) Adult & pupae in mango seed; (b) Damage to mango fruit. III. Literature Survey This section explains the different approaches carried out by the researchers in the area of image categorization and fruit diseases identification. Fruit disease identification can be seen as an instance of image categorization Sandeep S Musale, and Pradeep M Patil [1],have proposed Database Development of healthy Alphanso mangoes. To retain the overall mango fruit quality and to reduce postharvest losses during supply chain can be achieved by adopting suitable postharvest novel technologies involving image processing and pattern recognition techniques. They generated a soft tissue database of alphonso mangoes. Using digital X-ray machine. The database generated here contains 65% of defected mangoes whereas 35% healthy mangoes. Shiv Ram Dubey,Anand Singh Jalal,, GLA University Mathura, India[2] Adapted Approach for Fruit Disease Identification using Images. Have considered diseases of apple as a test case and evaluated for three types of apple diseases namely apple scab, apple blotch and apple rot. They used K- Means Clustering, Local Binary Pattern, Multi-class Support Vector Machine, Texture Classification. They consider two color-spaces (i.e. RGB and HSV color-space) and compared the performance of the system under these color spaces. The classification accuracy for the proposed solution is achieved up to 93%. En-Cheng Yang and team[3] have designed and developed methodology for detecting Early Infestations Caused by Oriental Fruit Fly (Bactrocera dorsalis) in Fruit using Non-Destructive Quarantine Technique. They tested the possibility of using X- rays to examine internal injuries of various fruit and measured performance of various fruits like apple, tomato, peach, guava, pitaya etc.applying global threshold value, this determined using single tailed Z- test. The critical value of z-test can b used as threshold value and then map Gradient image to binary image, Further morphological operator is applied to binary image to screen small and irrelevant spots. Vilas D. Sadegaonkar, Mr.Kiran H.Wagh[4]. The paper presents the recent development and application of image analysis and computer vision system in quality evaluation of products in the field of agricultural and food. There is a need for an efficient mango grading method to be developed. In this study, they proposed and implement 205

3 methodologies and algorithms that utilize digital image processing, content predicated analysis, and statistical analysis to determine the grade of local mango production. Payel G and Sunil CK[5] have designed and developed methodology for detecting Defect of internal break down causes browning or blackening of arils, which gives a typical foul smell if unnoticed. These internal defects cannot be identified from external appearance, which is a serious threat to the pomegranate processing industry as well as export resulting in rejections and quality reduction. X-ray inspection has a distinct advantage over other detection techniques, as it is non-destructive imaging of interior features of sample which detects internal defects. A continuous soft X-ray system with a semiconductor detector was used to detect the internal quality of the fruit. IV. PROPOSED WORK This section explains the proposed work. Figure 2 shows the architecture of this proposed work. Image Acquistion images are allowed to pre-process. Edges are detected using Edge detection algorithms. The percentage affected by stone weevil is identified. Then the sample is analysed and stone weevil is identified and percentage affected by the stone weevil is calculated classification and grading of samples is done based on the percentage of affected area in the sample. The detailed block diagram of adopted methodology is shown in Figure 2. V. METHODOLOGY 4.1 Image Acquisition Mango samples are collected both affected as well as healthy mangoes and images are scanned using digital x-ray machine. The Specifications with which X-ray exposure was taken are as: Voltage = 54 kv; Current = 100 ma; Exposure Time = 4 ms;d ~ 1 m. 4.2 Pre-processing: The next step is pre-processing of the image chosen for analysis. This module is designed to extract features of mango image. 4.3 Edge detection: The next step is to detect the edges of the sample.prewitt operator is used for edge detection in an image. Edges are calculated by using difference between corresponding pixel intensities of an image. Database Pre-process the Image Analyze the Image Algorithm1: Detecting edges of samples Start Input: scanned image of mango. Output: detect edges of the sample Step1: accept the image samples Step2:identify the edges of the give samples(mango) Step3:display the edges of the samples Stop Identifying the Edges Detecting the stone weevil Figure 2: Block diagram of proposed system First step is to collect the mango fruit and capture the image of the mango using digital X-ray. Scanned 4.4 Estimation of percentage of stone weevil affected area: The percentage of affected area is calculated by using equation (1) Percentage of healthy area = (healthy area/total data)*100, where healthy area is white pixels count in scanned image. Total area is total size of fruit. Percentage of affected area=100- percentage of healthy area. 4.5 Classification: Classification is the final step of this methodology, classification is done based on the percentage of disease affected area. Here mango fruit is classified into 5 categories: very lightly, lightly, moderately, 206

4 severely, uninfested mango. The algorithm used for the estimation of percentage of affected area and classification is given below. Algoritm2: estimation of percentage of affected area and classification. Input: Image samples Output: Percentage of affected area. Step1:Accept image samples Step2:Identify the edges using prewitt edge detection algorithm Step3: Estimate the percentage of affected area using equation(1) Step4if(percentage <1) display uninfested mango ; else if(percentage<=10) display 'Very Lightly infested else if(percentage <=30) display 'Lightly infested '; else if(percentage <=40) display moderately infested'; else display severely infested '; Stop. After this the obtained result is stored in database for future use or future reference. Figure 5: Infested portion displayed Figure 6 shows result comparisons of different edge detection algorithms applied to a mango fruit. Analysis shows that canny edge detection method gives better performance compared to other edge detection techniques. VI. RESULT The original image uploaded and its pre-processing and result is shown below. The result interface shows the percentage affected by stone weevil in mango fruit. The amount of stone weevil affected is displayed and result also gives the category into which the test sample falls is shown in Figure 4 and Figure 5. Figure 6: Algorithm comparisons VII. CONCLUSION Proposed system is an automated technique to estimate the percentage of stone weevil in mango fruit. System finds the percentage of stone weevil along with report generation that gives information about the fruit whether it is stone weevil affected or healthy fruit. Use of image processing makes it accurate and error free. All the algorithms used in this work are implemented using MATLAB 7.0. Figure4: Detecting edges and percentage affected. REFERENCES [1] Sandeep S Musale, and Pradeep M Patil Database Development of healthy Alphanso mangoes Int'l Journal of Advances in Agricultural & Environmental Engg. (IJAAEE) Vol. 1, Issue 1, ISSN EISSN ,

5 [2] Shiv Ram Dubey,Anand Singh Jalal. Adapted Approach for Fruit Disease Identification using Images GLA University Mathura, India,2011. [3] En-Cheng Yang, Man-Miao Yang*, Ling-Hsiu Liao, Wen- Yen Wu Non-Destructive Quarantine Technique- Potential Application of Using X-ray Images to Detect Early Infestations Caused by Oriental Fruit Fly (Bactrocera dorsalis) (Diptera: Tephritidae) in Fruit Formosan Entomol. 26: ,2006. [4] Vilas D. Sadegaonkar, Mr.Kiran H.Wagh Quality Inspection and Grading Of Mangoes by Computer Vision & Image Analysis ISSN : , Vol. 3, Issue 5, Sep-Oct 2013, pp ,2013 [5] Payel G and Sunil CK Quality Analysis of Pomegranate by Non-Destructive Soft X-Ray Method Indian Institute of Crop Processing Technology, Thanjavur, India. J Food Process Technol Volume 5,Issue 6, ISSN: JFPT,2014. [6] V. Ravindra and S. Shivashankar, Spongy tissue in Alphonso mango; A key evidence for the causative role of seed, Current Science, Vol- 91, NO-12, 25 December, [7] Tomas U. Ganiron Jr., Size Properties of Mangoes using Image Analysis, International Journal of Bio-Science and Bio-Technology vol 6, no 2, pp 31-42,