简介:在这份报纸,我们建议了能与深convolutional从食物图象识别盘子类型,食物成分,和煮的方法的一个多工系统神经网络。我们为每个班与至少500幅图象建立了不同食物的360个班的数据集。到数据的噪音,它是从因特网收集了的还原剂,孤立点图象通过与深convolutional特征训练的一个类的SVM被检测并且消除。我们同时训练了一个盘子标识符,一个煮的方法识别器,和一个多标签成分察觉者。他们在深网络体系结构分享一些低级的层。建议框架与手工制作的特征,和识别器和成分察觉者能被用于没在训练数据集被包括为用户提供引用信息的盘子的煮的方法比传统的方法显示出更高的精确性。
简介:Background:Theuseofnutritionalsupplements(NS)amongathletesiswidespread.However,littleisknownabouttherelationshipbetweennutritionaladequacyandNSusage.TheaimsofthisstudyweretoevaluatetheNSusageandtocomparethenutritionalintakefromfoodandprevalenceofmicronutrientinadequacy(PMI)betweenNSusersandnon-users.Methods:Portugueseathletesfrom13sportscompletedanNSusagequestionnaireandasemi-quantitativefood-frequencyquestionnaireassessinginformationovertheprevious12months.Theestimatedaveragerequirementcut-pointmethodwasusedtocalculatePMI.GenerallinearmodelswereusedtocomparenutritionalintakeandNSusage.Chi-squaredtestsandlogisticregressionwereperformedtostudy,respectively,relationshipsandassociationsbetweenPMIandNSusage.Results:Fromthe244athletes(66%males,13–37years),64%reportedNSusage.Afteradjustment,NSusersshowedahigherintakefromfood(p<0.05),foratleast1gender,forenergy,andfor7ofthe17studiednutrients.ThehighestPMIwereseenforvitaminsDandE,calcium,folate,andmagnesium.Afteradjustment,NSusers,irrespectiveofgender,reportedlowerPMIforcalcium(OR=0.28,95%CI:0.12–0.65),andfemaleusersformagnesium(OR=0.06,95%CI:0.00–0.98).Conclusion:AthletesusingNSreportedahighernutritionalintakefromfood,andalowerPMIforseveralnutrients.Perhaps,thosewhoweretakingNSwereprobablytheoneswhowouldleastbenefitfromit.
简介:Background:Theroleofwildfoodsincombatingproblemsoffoodshortageisparamount.However,existingapproachestocombatfoodinsecurityshockhavegenerallyfocusedonreducingvulnerabilityviaincreasingproductivityofdomesticatedfoods.Incontrast,approachesthatenhanceresiliencemainlythroughwildfoodsourceshavebeenlessfocused.ThisstudyexaminedthecontributionofwildfoodstohouseholdresiliencetofoodinsecurityinthegreenfaminebeltofEthiopia.Methods:Across-sectionalsurveyof220householdswasconductedusingastructuredquestionnaire,keyinformantinterviews,andsemi-participantobservations.FactoranalysiswasrunusingSPSStoanalyzedata.Correlationanalysiswasusedtoexaminethedirectionandstrengthofassociationbetweenwildfoodsandtheincomeandfoodaccess(IFA),alatentproxyindicatorofresilience.Cross-tabulationwasalsoruntodeterminetheproportionofhouseholdsineachethno-culturegroupundereachresiliencecategory.Results:Themeanamountofwildfoodsobtainedbyhouseholdswas156.61kgperhouseholdperannum.Thiswasabout5%and9%of,grossand,netfoodavailablefromallsourcesrespectively.Wildfoodscontributedwelltohouseholdresilienceasthefactorloading(Factor2=0.467)waslargeenoughandweresignificantlycorrelatedwithIFA(r=0.174).Wildvegetableswerethemostcollectedandconsumedtypeofwildfoodsconstituting52.4%oftotalamountofwildfoods.Thetotalamountofwildfoodswassmallerthanthatofdomesticatedsourcesoffood.Themajorityofhouseholds(38.6%)reported'reducedsourceofwildfoods'asareasonforthis.Smallerproportionoftheindigenous(11.2%)thanthenon-indigenous(34.1%)ethno-culturegroupreportedoneormorereasonsfortheirlowerlevelofdependenceonwildfoods.Conclusion:Fromthestudyweconcludedthatwildfoodshadimportantcontributiontohouseholds'resiliencetofoodshortagesandarelikelytocontinuetocontributeinthefuture,thisbeingmoretoindigenousthannon-indigenousethno-