EXTRACTION OF AYURVEDA HERBS AND BENEFIT USING DEEP LEARNING ALGORITHMS

International Research Journal of Engineering and Technology (IRJET)

Volume: 09 Issue: 04 | Apr 2022 www.irjet.net

e ISSN: 2395 0056

p ISSN: 2395 0072

EXTRACTION OF AYURVEDA HERBS AND BENEFIT USING DEEP LEARNING ALGORITHMS

1Assistant Professor. Department of Computer Science and Engineering, Sri Chandrasekharendra Saraswathi Viswa Mahavidyalaya, SCSVMV [Deemed to be University], Kanchipuram, Tamil Nadu

2,3Department of Computer Science and Engineering, Sri Chandrasekharendra Saraswathi Viswa Mahavidyalaya, SCSVMV [Deemed to be University], Kanchipuram, Tamil Nadu, India

***

Abstract: Ayurveda,Yoga,Unani,Siddha,andHomeopathy aresomeofIndia'straditionalmedicinalsystems.Ayurveda is effective in healing ailments without causing adverse effects.Medicinalplantsorherbsareregardedasavaluable resource for satisfying people's health care needs. It is necessarytopreserveanddigitiseinformationregardingthis therapeuticknowledge.Intheformofunstructuredtextual data, there have been a huge number of publications and articles on Ayurveda research. Text mining is utilised to provide a solution for dealing with such large amounts of unstructured data. With the exponential growth of text baseddata,findingthenecessaryinformationhasbecomea difficultchallenge.Theabilitytounderstandthesemanticsof document content is essential for assuring the quality of contentretrieval.

However,currentapproachesarediscoveringvariationsin textual categorization in order to improve classification accuracy,whichmayresultinafailuretocomprehenddata duringclassification.

Keywords: Ayurveda Herbs Images and details, Deep Learning,CNN,Densent121,Resnet50.

Introduction: Ayurveda is a traditional Indian medical systemthathasbeenpractisedsince2000B.C.Ayurvedais theSanskritwordfor"lifescience."Ayurvedaisasystemof medicine derived from nature. Over a thousand years of research,experimentation,anddocumentationofhundreds of plants, India's ancient sages have arrived at reliable findingsontheefficacyofvariousplantsandherbs.

AlthoughAyurveda'sefficacyforarangeofhumanproblems iswellknowninandaroundIndia,muchoftherestofthe world is unaware of the benefits that this unique Indian systemofmedicinemayprovide.Themajorityofayurvedic medicationshavenonegativesideeffectsorreactions.Many internalproblemsthataredeemedstubbornandincurable inothersystemsofmedicinecanbetreatedrationallywith Ayurveda. Ayurveda views lifeas a blend of senses, mind, body,andsoul.Asaresult,itisapparentthatAyurvedaisnot confinedtobodyorbodilyailments,butalsoencompasses spiritual, mental, and social well being. Vata, Pitta, and Kapha are three basic mind body types or doshas in

Ayurveda,andtheyrepresentdiversecombinationsofthe fiveelements:

ments air,ether,fire,water,andearth.Theyaremorelikely to develop anxiety, asthma, heart illness, nervous system issues,rheumatoidarthritis,andskinproblemsifvatadosha is the predominant life force. They are more likely to develop:Angerandbademotions,Crohn'sillness,heartburn afewhoursaftereating,highbloodpressure,andinfections ifpittadoshaisthepredominantlifeforce.Theyaremore likely to develop asthma and other lung illnesses, cancer, diabetes,nauseaaftereating,andobesityifthekaphadosha isthedominantlifeforce.Oneishealthywhentheseaspects areinbalance.Alldisordersareexcessesofoneormoreof these elements, and illness is defined as an imbalance of these elements. Ayurveda is a system of natural medicine thatusesnature'spowertobringpeoplebackintobalance.

The vast amount of data produced everywhere in today's digital world has resulted in a data intensive world. The internetconnectspeopleallovertheworld,allowingfordata sharing, a wide range of contact, and worldwide communication. Every person, as well as practically all businessesandprocedures,generatesdata.Thistextualdata isallaroundus,andit'sdevelopingatatremendouspace.

2) EXISTINGSYSTEMS:

This model emphasizes an existing method that which is designedusingthesomeoftheMachinelearningapplication throughthedatasetin.CSVfileandusingSVM,DecisionTree andRandomforestmethods

2.1) PROPOSED SYSTEM:

Inpurposedmethodweareperformingtheclassification ofeithertheusingAYURVEDAHERBS&BENEFITSimages ConvolutionNeuralNetwork(CNN)ofdeeplearningalong withtheTransferlearningmethods.Asimageanalysisbased approachesforimageclassificationofherbanditsbenefits. Hence, properclassificationisimportantfortheproperusing anditsdetailsofplantsthatwhichwillbepossiblebyusing ourproposedmethod.Blockdiagramofproposedmethodis shownbelow.

Factor value:

| ISO 9001:2008

HighAccuracy

Performance

Classificationis

Lossis

REQUIREMENTS

ISSN:

0056

0072

System

User

1. System:

1.1CreateDataset:

The dataset containing images of the Ayurvedic Herbsimagessetsandclassificationofherbsdetails i.e.,normalaretobeclassifiedissplitintotraining andtestingdatasetwiththetestsizeof30 20%.

1.2Pre processing:

Resizingandreshapingtheimagesintoappropriate formattotrainourmodel.

1.3Training:

Use the pre processed training dataset is used to train our model using CNN Deep learning with a long of Resnet50, Denset121 transfer learning methods.

1.4Classification:

Theresultsofourmodelaredisplayofplantdisease classificationimagesareeitherwithdifferentlabels andproperties

2. User:

2.1UploadImage

Theuserhastouploadanimagewhichneedstobe classified.

2.2ViewResults

Theclassifiedimageanddetailsresultsareviewed byuser.

Steps and implementation:

First,wetaketheHerbsimages

We prepare the images the Pre processing for training

Afterthatwesplitthedata70% 30%or80% 20% fortestingprocess

After splitting data train with model CNN, Densnet121,Resnet50

ThenclassificationtheimagestotesttheHerbitem namesanditsproperties

Home: In our project, we are classifying the presence of AyurvedaHerbsClassificationandProperties,withthehelp ofdeeplearning.

Upload Image:

Here the images can be uploaded those which are to be classified.

Classified output:

The uploaded image is classified as the Alovera and its Properties.

Model choosing:

Herethemodelcanbeselected,bywhichtheimageistobe classified.

Classified output:

The uploaded image is classified as the AIMA and its properties.

Classified output:

The uploaded image is classified as the GUGGUL and its properties.

Classified output:

The uploaded image is classified as the HARIDRA and its properties.

Classified output:

TheuploadedimageisclassifiedasthePEPPERandits properties.

and Technology (IRJET)

e ISSN: 2395 0056

ISSN: 2395 0072

intotheareaoftextminingwithouthavinganypreviousback groundincomputationallinguistics.

S.A.Dahanukar,R.A.Kulkarni,N.N.Rege

PharmacologyOfMedicinalPlantsAndNaturalProducts

Hence, networking of the various research activities carriedoutbydifferentscientistshasnowbecometheneed ofthehourinordertoprovideinformationaboutandaccess to research work done in different laboratories of the country.

Dileep,M.R.,&Pournami,P.N.

Deep LearningApproachforClassification of Medicinal Plants

The graph narrates the progress of accuracy and loss functions during the training process. A five fold cross validation is performed on AyurLeaf dataset, maximum accuracyachievedforasinglerunoutoffiveconsecutiveruns is98.46%.

Gopal,A.,PrudhveeswarReddy,S.,&Gayatri,V. Classificationofselectedmedicinalplantsleafusingimage processing

Classified output:

TheuploadedimageisclassifiedastheNEEMandits properties.

4

. SYSTEMARCHICTURE

5. Literature survey

CambridgeUniv.Press,2007

RonenFeldmanandJamesSanger

AdvancedApproachestoAnalyzingUnstructuredData

Bybeingself contained,withseveralchaptersaddressing indetailthemaintopicsrelevanttotextmining,thebookis highlyrecommendedtothosewhowouldliketostartdelving

Factor value:

Theboundary basedfeatures,momentfeatures&colour oftheleavesareusedforthepurposeofidentificationofleaf varieties.Suchautomatedclassificationmechanismscanbe useful for efficient classification of plant leaf species. The accuracyofthecurrentmethodwasfoundtobe92%.

6. CONCLUSION:

Inthisprojectwehavesuccessfullyclassifiedtheimagesof Identification of Ayurveda Herbs Classification and its properties to use for diseases. how to use for medical purposes with the help of deep learning and Transfer learningmethodslikeCNN,Densnet121,Resnet50.

7. References

[1]A.CamargoandJ.S.Smith,“Animage processingbased algorithm to automatically identify plant disease visual symptoms,” Biosystems Engineering, vol.102, pp.9 21, January2009.

[2] J.S. Cope, D. Corney, J.Y. Clark, P. Remagnino, and P. Wilkin, “Plant species identification using digital morphometrics: A review,” Expert Systems with Applications,vol.39,pp.7562 7573,June2012.

[3]J.GarciaandA.Barbedo,“Usingdigitalimageprocessing forcountingwhitefliesonsoybeanleaves,”JournalofAsia PacificEntomology,vol.17,pp.685 694,December2014.

9001:2008

International Research Journal of Engineering and Technology (IRJET)

Volume: 09 Issue: 04 | Apr 2022 www.irjet.net

[4]A.Gongal,S.Amatya,M.Karkee,Q.Zhang,andK.Lewis, “Sensorsandsystemsforfruitdetectionandlocalization:A review,”ComputersandElectronicsinAgriculture,vol.116, pp.8 19,August2015.

[5]I.Goodfellow,Y.Bengio,andA.Courville,DeepLearning, MITPress,2016.

[6]J.HemmingandT.Rath,“Computer vision basedweed identification under field conditions using controlled lighting,” Journal of Agricultural Engineering Research, vol.78,no.3,pp.233 243,March 2001

Factor value: 7.529 | ISO 9001:2008

e ISSN: 2395 0056

p ISSN: 2395 0072