“ARTIFICIAL INTELLIGENCE BASED CALORIE ESTIMATOR FOR FRUIT”

Tejswini Balpande1, Nikita Dhothkar2, Heena Satpute3, Namrata Durbude4, Vijay.V.Chakole5
Electronics Department, K.D.K.C.E, Nagpur
1tejubalpande@gmail.com,2Nikitadhotkar3@rediffmail.com,
3heenasatpute2019@gmail.com,4durbudenamrata@gmail.com

Abstract:

This project is used to identify the fruits from image is the interesting field with various application The main objective of this project is to detection of fruits and estimate calories using the image acquired by camera. This study proposes a methodology for automatic fruit recognition and calorie measurement using artificial intelligence. The identifying the fruits and calories from an image is quite an interesting field with various applications. Since fruit monitoring plays an important role in health-related problems, it is becoming more essential in our day-to-day life. For this purpose an recognition for fruit is presented by considering its shape, color, size and texture characteristics. The main goal of our research is to enhance and improve the accuracy of dietary assessment by analyzing the fruit images captured by using mobile devices (e.g. smart phone). It requires mobile application to have the internet connection, but it’s independent from your programming language choice and resources limitations which is important for mobile device.

Keywords – Artificial Intelligence,), Calorie measurement, Fruit image recognition.

Ⅰ. INTRODUCTION

The process is improved in such a way that it takes less time to complete. Automation is plays important role in day to day life. Their main source of income is agriculture. Exporting of fresh fruits is increase day to day from India. In this time the people are very serious about their health; they desired only for fresh and good quality fruit. In this project has fruit detection and recognition calories app can count the calories of fruits from photos on people. The app that uses advance image recognition technology, it will able to establish any fruits qualities that capture in photos and associate a calorie quality to each items. Recently, smart applications for mobile devices such as Android phones and iPhone, have increased tremendously. So in this we use android studio for detection of fruit quality and finding the calories from an image captured by using mobile camera which is act as a hardware part. One of the major goal of fruit image processing is to retrieve calorie information from the given fruit image. In addition, automatic fruit recognition is beneficial to health care related applications, such as obesity management. They are capable of processing a real time application. Since the present smartphones can handle the high fruit image quality and focused on developing real time applications which capture image then scan and automatically can detect the good quality of fruits.

II. AIM AND OBJECTIVE

Aim of the project

Aim of the Project is automatically detect the fruits using an Artificial Intelligence based on algorithm which can detect fruits items from image acquired by using camera.

Objective of the project

Objective of our project is to identify the fruits based on their quality. Our objective is to develop an efficient fruit processing system to calculate the calorie and nutrition of fruit.

III. REQUIREMENTS

The requirements and constraints that our application was to meet during implementation. The Functional requirements detail what precisely our system does, while the Non-Functional requirements deal with how it performs. To illustrate each requirement’s priority, all are enumerated within their distinct sections. We also discuss the Design Constraints that served as limiting factors on the implementation of this project.[5]

Functional

1. Application will make use of its devices camera to take photos for analysis.
2. The application will correctly identify fruits in a given image.
3. The application will calories per serving of identified food.
4. The application will store users’ fruit history and associated data in a Fruit History.
5. The application will allow users to make accounts and securely sign in.

Non-functional

1. The application will be accurate in its predictions.
2. The application will be easy to use.
3. The application will be efficient and lightweight.
4. The application will return results within a reasonable time-frame.

Design Constraints

1. The application will run on Android mobile devices.
2. The application will only function while connected to the Internet.

Ⅳ. RELATED WORK

We are going to show how ever fruit detection can take place with use of above steps for every detected fruit portion, a feature extraction method must be used. Later the research has concentrate on collecting the data set which helps to monitor diets. The set of local features are also extracted on using the classifiers. It is shows that every image goes from filtering operation. To calculate surface area for fruit portion we formed to superimpose a grid of squares on to the image segment so each square contain equal no. of pixel and , equal number of area. In this paper we make a measurement method that measures the amount of calories from fruit image by measuring the volume of fruit portion from image and nutrition facts table. In this paper we proposed artificial intelligence to address the fruit image recognition problem. Specifically we proposed convolution neural networks based algorithm. In this classification of framework that takes advantages of the user speech input to increase the fruit recognition process. In this paper we involved machine learned features with deep learning based method to achieve a much higher accuracy.

Ⅴ. METHODALOGY

Many papers have been presented to solve the problems of fruit recognition. The work done of fruit detection system is first initiated with many classes. The first related research area is tehnology solution for enhancing the accuracy of dietary measurement.we used the android studio software for fruit detection. By using the mobile app we can measure the calories of fruits. The several app have an impeoved automation. In this project we are used the different classes which are as follows:

1. Background class we will implement the searching optimization.
2. Permission required.
3. Launching activity.
4. Calories finder and saved.
5. Image recougnization.
6. Image recougnize with finder.

Ⅵ. WORKING

Figure 1.Potential Action Available to User[5]

Register
Actors: User
Goal: User makes an account for the application.
Precondition: User has installed the application.
Post condition: User added to user list.
Exception: Entering in a currently used username will cause an error and force the user to choose a new one.

Sign In
Actors: User
Goal: User enters username and password into application to gain access to their account.
Precondition: User must have an account.
Post condition: User gains access to account information and application functionality.
Exception: If the user enters the wrong use name or password, an error will be returned that either the user name or password was incorrect.

Take Picture
Actors: User
Goal: User takes picture of their fruits utilizing local phone camera.
Precondition: User is registered and signed in to their account.
Post condition: Detection results will be shown to the user after analysis from the server.
Exception: If the user does not allow camera access, an error will be sent to the user with a direction to allow access to the camera.

Fruit detection and Carries Information
Actors: User
Goal: User click in food log and save fruit detection and calories information.
Precondition: user hold the camera while fruit and calories were detected.
Post condition: Get all the information about fruit detection and calories.
Exception: If the user not get proper information about fruit detection and calories processing then user get back to the camera otherwise done logging.

Add Fruit History
Actors: User
Goal: User adds the detected fruit item to their fruit history.
Precondition: The picture has been returned to the user and labeled by the database.
Post condition: The fruit history will have been updated with the new value on the date the entry was added.
Exception: If no items were detected the user will be unable to add anything.

Review Fruit History
Actors: Admin
Goal: User can audit passages from current/previous days they have entered information.
Precondition: User must have added data to the fruit history before hand.
Post condition: N/A.
Exception: N/A.

Ⅶ. TECHNOLOGY USED

In this project we are used the following technology:

1. Android: Android is an open source and Linux-based operating system language for mobile devices such as smartphones and computers. It will be easy to develop an application on it. Hence in this project we are going to use android application for the front end.
2. Java: We are developing an Android, java is one of the best language for application development. The official language for Android development is java. Most of the part of android is written in java. Hence java is a compiler language which can be used for back end.
3. Firebase: In this the firebase is only useful for Google account authentication in our application.

Ⅷ. ARTIFICIAL INTELLIGENCE [AI]

It is a branch of computer science by which we can create intelligent machine which can behave like a human think like a human and to make decision. It endeavors not fair to get it but moreover to construct brilliantly substances. John Mccarthy says that, “The Science and Engineering of making shrewdly computer programs.’’ The goal of AI is creating a system or machine which implement human intelligence means it must be able to

1. Undertake
2. Think
3. Learn
4. Behave like human

Each perspective of learning or any other include of insights can in guideline be so accurately portrayed that a machine can be made to mimic it. An endeavor will be made to discover how to form machine use dialect, from reflection and ideas fathom sorts of issues presently saved for people and move forward themselves.

Ⅸ. CAMERA REVIEW

Ⅹ. IMAGE PROCESSING

Image processing could be a strategy to perform a few operations on an picture , in arrange to urge an improved picture or to extricate a few valuable data from it.. It is type of signal processing is among rapidly growing technologies It shapes center inquire about range inside engineering and computer science disciplines as well. Image processing basically includes the following three steps:

1. Importing the image via acquisition tools.
2. Analyzing and manipulating the image.
3. Output in which result can be changed picture or report that’s based on picture analysis.

Ⅺ. CONCLUSION

We proposed a method that estimate the amount of calories from a fruit image is analyzed. The AI technology can consequently identify the natural product securing a genuine world self-centered picture with sensible precision, lessening the information handling and protection concern. In this regard, we succeeded, as we made an application that fulfilled these expectations and passed the tests we set out for the system. This section details what we learned from this project, as well as what we think would be substantial improvements to the application.

Ⅻ. REFERENCES

1. Yuzhen Lu ,”Food Image Recognition By Using Convolutional Neural Networks (CNNs)”,Department of Biosystem And Agricultural Engineering, Michigan State University, East Lansing, MI 48824, USA; 2019.
2. David J.Attokaren, Ian G.Fernandes, A.Sriram, Y.V Srinivasa Murthy,” Food Classification From Images Using Convolution Neural Networks”, and Shashidhar G. Koolagudi department of CSE, IEEE 2017
3. Kiran Ambhore, Prof.N.A. Dawande,” Measuring Calories And Nutrition From Food Image”, IEEE 2016
4. Marios M.Anthimopoulos, Member, IEEE, Lauro Gianola, Luca Scarnato,Peter Diem,”A Food Recognition System For Dibetic Patients Based On An Optimise Bag-Of-Features Model”, And Stavroula G. Mougiakakou,member,IEEE,2014
5. Hoff, Stephen; Jaffurs, Patterson; Enriquez, Michael; and Wilde, Quintin, “Snap-n-Snack: a Food Image Recognition Application” (2018).Computer Engineering Senior Theses. 121. https://scholarcommons.scu.edu/cseng_senior/121
6. Muhammad farooq, and edward sazonov, “Accelerometer-based detection of food intake in free-living Individuals”, IEEE sensors journal., vol. 18, no. 9, pp. 3752–3758, 2018.
7. Diptee Kumbhar, and Sarita Patil, “Mobile cloud based system recognizing nutrition and freshness of food Image”, energy, communication, data analytics and soft computing, IEEE conference on, pp. 709-714, 2017.
8. Martin, CK, Nicklas, T, Gunturk, B et al. (2014) Measuring food intake with digital photography. J Hum Nutr Diet 27, Suppl. 1, 72–81.
9. Stumbo, PJ (2013) New technology in dietary assessment: a review of digital methods in improving food record accuracy. Proc Nutr Soc 72, 70–76.
10. Boushey, CJ, Kerr, DA, Wright, J et al. (2009) Use of technology in children’s dietary assessment. Eur J Clin Nutr 63, Suppl. 1, S50–S57.
11. S.Bianco,G.Ciocca,P.Napoletano,and R.Schettini,“Aninteractivetool formanual,semi-automatic andautomatic video annotation,”Comput.Vis. Image Underst., vol. 131, pp. 88–99, 2015.
12. Food Recognition and Calorie Extraction using Bag-of- SURF and Spatial Pyramid Matching Methods Hattarki.Pooja1 , Prof. S.A.Madival2 1 P.G Student, Dept CSE Appa Institute of Engineering and Technology, Kalaburagi, Karnataka, INDIA.
13. M. Bosch, F. Zhu, N. Khanna, C. Boushey, and E. Delp, “Combining global and local features for food identification in dietary assessment,” in Proc. 18th IEEE Int. Conf. Image Process., 2011, pp. 1789–1792.
14. M. M. Anthimopoulos, L. Gianola, L. Scarnato, P. Diem, and S. G. Mougiakakou, “A food recognition system for diabetic patients based on an optimized bag-of-features model,” IEEE J. Biomed. Health Informat., vol. 18, no. 4, pp. 1261–1271, Jul. 2014.
15. Y. He, C. Xu, N. Khanna, C. Boushey, and E. Delp, “Analysis of food images: Features and classification,” in Proc. IEEE Int. Conf. Image Process., 2014, pp. 2744–2748.
16. G. Farinella, M. Moltisanti, and S. Battiato, “Classifying food images represented as bag of textons,” in Proc. IEEE Int. Conf. Image Process., 2014, pp. 5212–5216.
17. V. Bettadapura, E. Thomaz, A. Parnami, G. Abowd, and I. Essa, “Leveraging context to support automated food recognition in restaurants,” in Proc. IEEE Winter Conf. Appl. Comput. Vis., 2015, pp. 580–587.
18. W. Zhang, D. Zhao, W. Gong, Z. Li, Q. Lu, and S. Yang, “Food image recognition with convolutional neural networks,” in Proc. IEEE 12th Int. Conf. Ubiquitous Intell. Comput. IEEE 12th Int. Conf. Auton. Trusted Comput. IEEE 15th Int. Conf. Scalable Comput. Commun. Assoc. Workshops, 2015, pp. 690–693.
19. O. Beijbom, N. Joshi, D. Morris, S. Saponas, and S. Khullar, “Menumatch: Restaurant-specific food logging from images,” in Proc. IEEE Winter Conf. Appl. Comput. Vis., 2015, pp. 844–851.
20. P. Pouladzadeh, S. Shirmohammadi, and R. Al-Maghrabi, “Measuring calorie and nutrition from food image,” IEEE Trans. Instrum. Meas., vol. 63, no. 8, pp. 1947–1956, Aug. 2014.
21. Y. He, C. Xu, N. Khanna, C. Boushey, and E. Delp, “Food image analysis: Segmentation, identification and weight estimation,” in Proc. IEEE Int. Conf. Multimedia Expo, 2013, pp. 1–6.