Please use this identifier to cite or link to this item: https://dspace.univ-ouargla.dz/jspui/handle/123456789/30896
Title: A healthcare system using deep learning
Authors: AIADI, Oussama
BENNOUH, Radja
Keywords: deep learning
X-ray
datase
ensemble learning
Issue Date: 2022
Abstract: Classification of images is useful for extracting different useful information, for example, classification of animals (cat, dog...etc), classification of plants, classification of X-ray images to tell the injured from the uninfected, all these things are useful, in this thesis we are interested in developing A system for identifying the type of insect from the stinger using deep learning. The proposed approach is to try several models of the convolution neural network (CNN) and see the classification ability of each model and then apply the learning set method to improve the system's classification accuracy. We run experiments on our own dataset created with careful research and effort and the help of an expert in the medical field and guidance to know the features of each bite
عٌذ تص فٍُ انصىس يف ذٍا لاستخشاج يعهىياخ يف ذٍج يختهفح ،عهى سث مٍ ان ثًال ، تص فٍُ انح ىٍا اَخ )قط، كهة ....انخ(، تص فٍُ ان ثُاتاخ، تص فٍُ صىس اشعح ص حٍٍُ ن عًشفح ان صًاب ي غ شٍ ان صًاب ، كم هز الايىس يف ذٍج ، ف هز الاطشوحح حَ يهت ىً تتطى شٌ ظَاو نهتعشف عهى ىَع انحششج ي اثش انهزغح تاستخذاو انتعهى انع قًٍ . ان هُح ان قًتشذ هى تدشتح عذج اًَرج ي انشثكح انعصث حٍ الانتفاف حٍ ويعشفىح قذسج تص فٍُ كم ىًَرج وي ثى تطث قٍ طش قٌح يد ىًعح انتعهى نتحس دقح ان ظُاو ف انتص فٍُ حَ دَشي انتداسب عهى يد ىًعح انث اٍ اَخ انخاصح ت اُ تى ا شَاؤها ت دًهىد وتحث دق قٍ والاستعا حَ تخث شٍ ف ان دًال انطث وتىخ هٍ اً ن عًشفح ي زٍاخ كم نزغح.
Description: industrial computer science
URI: https://dspace.univ-ouargla.dz/jspui/handle/123456789/30896
Appears in Collections:Département d'informatique et technologie de l'information - Master

Files in This Item:
File Description SizeFormat 
BENNOUH.pdf.OK.PDFindustrial computer science1,63 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.