Hybrid Approach for apple fruit disease detection, yield estimation and grading using YOLOv5

International Research Journal of Engineering and Technology (IRJET) e ISSN: 2395 0056

Volume: 09 Issue: 07 | July 2022 www.irjet.net p ISSN: 2395 0072

Hybrid Approach for apple fruit disease detection, yield estimation and grading using YOLOv5

1Associate Professor

2MTech in Digital Communication Dept. of ECE Dept. of ECE

BMS College of Engineering, Bengaluru

Abstract - India comes in behind China as the world's second largest producer of fruits and vegetables. A farmer must carry out a variety of duties, including yield detection, grading, and disease detection. To automatically identify disease symptoms as soon as they arise on developing fruits, early detection of fruit illnesses is crucial. Fruit infections can result in significant productivity and quality losses during harvest. In this study, the hybrid task of disease identification, yield estimation, and grading of apple fruit is carried out by YOLOv5. The proposed system's output is assessed using the measure mean Average Precision (mAP).

Key Words: YOLO,BoundingBox,ResidualBlocks

1.INTRODUCTION

Theapple(Maluspumila),whichranksfourthamongfruits producedworldwidebehindthebanana,orange,andgrape, isthemostsignificanttemperatefruitcommercially.Indiais the world's second largest producer of apples. Approximately 70% of Jammu and Kashmir's total population is dependent on agriculture, either directly or indirectly. Jammu and Kashmir's two main exports are apples and walnuts; the state contributes around 75% of India's apple crop. Fruit diseases have a severe impact on globalagriculturalindustryproductivityandfinanciallosses. Thetraditionalmethodforidentifyinganddiagnosingfruit illnessesreliessolelyonexpertobservationwiththeunaided eye.Duetotheiravailability'sremotelocations,consulting specialists can be costly and time consuming in some underdevelopednations.

Toautomaticallyidentifydiseasesymptomsassoonasthey emerge on developing fruits, automatic fruit disease detection is crucial. Apple fruit infections can result in significant yield and quality losses during harvest. It is essentialtounderstandwhatisbeingobservedtodetermine whatcontrolmeasurestousethenextyeartopreventlosses. Therearesomeillnessesthatspreadtothetree'sbranches, leaves,andotherparts,includingthetwigs.AsseeninFig.1, applescab,applerot,andappleblotcharesomeprevalent diseasesofapplefruits.Applescabsappearascorky,grey,or browndots.Infectionsthatcauseapplerotresultincircular, slightlydepressedbrownorblackpatchesthatmayhavea crimson halo. A fungal disease called apple blotch causes

BMS College of Engineering, Bengaluru ***

dark, wavy, or lobed margins on the fruit's surface. Producing an estimated count of crops is called fruit counting,whichisoftenreferredtoascropyieldestimates. Knowingtheyieldenablesfarmerstomanageresourcesfor harvesting wisely, prepare storage according to crop quantities, and develop more effective plans for harvest routesandpackaging.Automatingthisprocedurewouldbe quitehelpfulbecauseitistime consumingtosendpersonnel outintothefieldtocountthingsandbecausetheestimates mustbequickandprecise.Afterharvest,gradingthefruitsis acrucialpartofpost harvestmanagement.

(a)

(b)

(c)

Fig.1(a)AppleScab(b)Appleblotch(c)AppleRot

Fruit grading is the process of evaluating and classifying variousfruitclassesorstandards.Differentcategoriescanbe definedbysize,color,orevensoftnessorfaultlevel.After harvest,gradingthefruitsis a crucial partofpost harvest management.Fruitsandvegetablesareratedbasedontheir weight,size,color,form,specificgravity,andlackofillness, allofwhichdependontheagroclimaticconditions.Manual

International Research Journal of Engineering and Technology (IRJET) e ISSN: 2395 0056

Volume: 09 Issue: 07 | July 2022 www.irjet.net p ISSN: 2395 0072

gradingandsizegradingarethetwomostcommonwaysof fruit grading. Image analysis employs deep learning to addresstheseproblems.Deeplearningmodelsarenotonly very accurate, but also very adaptable. All the tasks are addressedbyYOLOV5model.

The YOLOv5 model is mostly appropriate for mobile and embedded devices because to its basic network structure. Additionally, this compact model detects quickly and requiresminimaltrainingtime.Usingthemakesense.aitool these images have been enhanced and annotated and the trained YOLOv5 model, identifying the objectsand determining the class to which they belong are the two aspectsoftherecognitionprocess.

2. PROPOSED SYSTEM

You Only Look Once is known by the acronym YOLO. This algorithmidentifiesandfindsdifferentthingsinapicture(in real time).Theclassprobabilitiesofthediscoveredphotos are provided by the object identification process in YOLO, whichiscarriedoutasaregressionproblem.Convolutional neural networks (CNN) are used by the YOLO method to recognize items instantly. The approach just needs one forward propagation through a neural network to detect objects,asthenamewouldimply.Thisindicatesthatasingle algorithmrunisusedtoperformpredictionthroughoutthe fullimage.Multipleclassprobabilitiesandboundingboxes aresimultaneouslypredictedusingtheCNN.

The following justifications make the YOLO algorithm crucial:

Speed: Becausethisalgorithmcanpredict objectsinreal time,itincreasesthespeedofdetection.

High accuracy: TheYOLOpredictionmethodyieldsprecise findingswithfewbackgroundmistakes.

Thealgorithmhasgreatlearningcapabilitiesthatallowitto pick uponobjectrepresentationsandusethemforobject detection.TheYOLOalgorithmemploysthefollowingthree methods:

Residualblocks

Boundingboxregression

IntersectionOverUnion(IOU)

Residual blocks:

First,theimageisdividedintovariousgrids.Eachgridhasa dimensionofSxS.

Fig.2Imagedividedintogrids

Bounding box regression:

A bounding box is a graphic that shows the location of an objectinaphotograph.Thefollowingascribesareincluded ineachboundingboxoftheimage:

Length(bw)

Dimensions(bh)

Class(forexample,apple,banana,tomato,andsoon) which isdenotedbytheletterc.

Boundingboxplace(bx,by)

Fig.3BoundingBox

YOLO uses a lone bounding box regression to predict the height,width,center,andclassofobjects.

Intersection Over Union:

Box overlapping is described by the object detection phenomenaknownasintersectionoverunion(IOU).IOUis usedbyYOLOtocreateanoutputboxthatproperlyencircles theitems.Thepredictedboundingboxesandtheirconfidence scoresaretheresponsibilityofeachgridcell.Iftheprojected boundingboxandtheactualboxmatch,theIOUisequalto1. Boundingboxesthatarenotequivalenttotheactualboxare eliminatedbythisapproach.

International Research Journal of Engineering and Technology (IRJET) e ISSN: 2395 0056

Volume: 09 Issue: 07 | July 2022 www.irjet.net p ISSN: 2395 0072

Neck Model:

NeckstyleusingtheFeaturePyramidsNetwork(FPN) based PANetonYOLOv5.ThePANetmodelisutilisedtosupportthe model'sgoodobjectscalinggeneralisation.Whenidentifying thesamethinginvarioussizesandscales,thisisincredibly helpful.

Head Model:

Fig.4Groundtruthboxandpredictedbox

Fig.4displaystwoboundingboxes,oneinredandtheother in yellow. The genuine box is represented by the red box, whiletheexpectedboxisrepresentedbytheyellowbox.The IntersectionOverUnion(IOU)techniqueisusedtoassessthis performance.TheIOUhelpsinfiguringoutwhetheraplaceis home to an object. As demonstrated below, the IOU is discovered by dividing the two boxes' association and convergencespacesbyeachother.TheIOUincreasesasthe outlookgetsbetter.

TheBackboneModel,theNeckModel,andtheHeadModel are the three components of the YOLO v5 architecture as showninFig5

Thefinaldetectionisperformedbytheheadmodel,which applies anchor boxes to the features and generates a final outputvectorwithboundingboxes,objectivityscores,and class probabilities. Leaky ReLU and Sigmoid are the activationfunctionsusedbyYOLOv5.Inthemiddlelayeror buriedlayers,theLeakyReLUactivationfunctionisutilised, andSigmoidisusedinthefinaldetectionlayer.Theartificial neuralnetwork'sactivationfunctiondetermineshowmany input weights and biases are needed to activate and deactivateneurons.

3. RESULTS AND DISCUSSION

The dataset used in the model is 1537 images which are dividedintotrainingandtestingdata.80%oftheimagesare trainingdatasetand20%areusedfortesting.

1.Building a Dataset

Using the makesense.ai tool, which tags things with rectanglesasseeninFig.6,datalabellingisdonemanually.

Fig.5YOLOv5architecture

Backbone Model:

Whengivenaninputimage,thebackbonemodelisutilisedto extractkeyfeatures.TheCSP(CrossStagePartial)Network servesastheframework forYOLOv5'sextractionofuseful characteristicsfromtheinputimage.

Fig.6Labellingusingmakesense.ai

2.Training Dataset

The primary framework is the YOLOv5 repository. All YOLOv5repositoriesanddependenciesmustbemet.1537 tagged images were used for the facial recognition sub datasetsystem'straining.Thetrainingwascarriedoutfor 300 epochs. The results following training of apple yield estimationaredisplayedinFig.7 intermsofF1curve,PR curveandconfusionmatrix.

International Research Journal of Engineering and Technology (IRJET) e ISSN: 2395 0056

Volume: 09 Issue: 07 | July 2022 www.irjet.net p ISSN: 2395 0072

Fig.7(a)F1curveforappleyieldestimation

Fig.8(a)F1curveforapplediseasedetection

Fig.

Fig.7(b)PRcurveforappleyieldestimation

Fig.8(b)PRcurveforapplediseasedetection

7

(c)Confusionmatrixforappleyieldestimation

Theresultsfollowingtrainingofapplediseasedetectionare displayedinFig.8

Fig.8(c)Confusionmatrixforapplediseasedetection

According to the training findings, the level of minimised return is attained when the graph begins to resemble an elbow (elbow technique) after 300epochs at mA@0.5, mAP@0.5:0.95,Precision,andRecall.Figure9displaysthe findings.

2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified

International Research Journal of Engineering and Technology (IRJET) e ISSN: 2395 0056

Volume: 09 Issue: 07 | July 2022 www.irjet.net p ISSN: 2395 0072

Accordingtothesizeofthefruit,applesaregraded.Thefruit isgivenanAorBgradebasedonathresholdthathasbeen establishedasshowninFig.10andFig.11.

Fig.12Applerot

Fig.9Resultsoftheproposedmodel

duringtraining

Aftertesting,theoutputimageswillincludeboundingboxes andanaccuracy%.

Fig.10Yieldestimationalongwiththegrade

Fig.11Yieldestimationalongwiththegrade

Fig.13AppleScab

ThesameYOLOv5model istrainedand tested for disease detectionandthemodelcanidentifythescabandrotdisease asshowninFig.12andFig.13.

4. CONCLUSION

Theworkisahybridapproachfordiseasedetection,yield estimationandgradingofanapplefruitusingYOLOv5.The images of the fruits used are collected from Google and Kaggle. Theproposedmodelhasresultedwith99%mAP. Theapproacheasestheworkofafarmerbyperformingall the major tasks of pre and post harvesting with high accuracy. It would simply be necessary to fine tune the object detector for our future work to test the model on differentfruits,suchoranges.

REFERENCES

[1] R. S. Latha et al., "Fruits and Vegetables Recognition using YOLO," 2022 International Conference on Computer Communication and Informatics (ICCCI), 2022,pp.1 6,doi:10.1109/ICCCI54379.2022.9740820.

International Research Journal of Engineering and Technology (IRJET) e ISSN: 2395 0056

Volume: 09 Issue: 07 | July 2022 www.irjet.net p ISSN: 2395 0072

[2] K. R. B. Legaspi, N. W. S. Sison and J. F. Villaverde, "Detection and Classification of Whiteflies and Fruit FliesUsingYOLO,"202113thInternationalConference on Computer and Automation Engineering (ICCAE), 2021,pp.1 4,doi:10.1109/ICCAE51876.2021.9426129.

[3] W. Yijing, Y. Yi, W. Xue fen, C. Jian and L. Xinyun, "Fig Fruit Recognition Method Based on YOLO v4 Deep Learning," 2021 18th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI CON), 2021, pp. 303 306, doi: 10.1109/ECTI CON51831.2021.9454904.

[4] Y.Osman,R.DennisandK.Elgazzar,"YieldEstimation using Deep Learning for Precision Agriculture," 2021 IEEE7thWorldForumonInternetofThings(WF IoT), 2021, pp. 542 550, doi: 10.1109/WF IoT51360.2021.9595143.

[5] H.Chopra et al.,"EfficientFruitGrading SystemUsing SpectrophotometryandMachineLearningApproaches," in IEEE Sensors Journal, vol. 21, no. 14, pp. 16162 16169, 15 July15, 2021, doi: 10.1109/JSEN.2021.3075465.

[6] C. Liu, Y. Tao, J. Liang, K. Li and Y. Chen, "Object Detection Based on YOLO Network," 2018 IEEE 4th InformationTechnologyandMechatronicsEngineering Conference (ITOEC), 2018, pp. 799 803, doi: 10.1109/ITOEC.2018.8740604.

[7] Y. Tian, G. Yang, Z. Wang, H. Wang, E. Li and Z. Liang, "Apple detection during different growth stages in orchards using the improved YOLO V3 model", Computers and electronics in agriculture, vol. 157,pp.417 426,2019.

2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal