50 research outputs found
Data Augmentation through Background Removal for Apple Leaf Disease Classification Using the MobileNetV2 Model
The advances in computer vision made possible by deep learning technology are increasingly being used in precision agriculture to automate the detection and classification of plant diseases. Symptoms of plant diseases are often seen on their leaves. The leaf images in existing datasets have been collected either under controlled conditions or in the field. The majority of previous studies have focused on identifying leaf diseases using images captured in controlled laboratory settings, often achieving high performance. However, methods aimed at detecting and classifying leaf diseases in field images have generally exhibited lower performance. The objective of this study is to evaluate the impact of a data augmentation approach that involves removing complex backgrounds from leaf images on the classification performance of apple leaf diseases in images captured under real world conditions. To achieve this objective, the lightweight pre-trained MobileNetV2 deep learning model was fine-tuned and subsequently used to evaluate the impact of expanding the training dataset with background-removed images on classification performance. Experimental results show that this augmentation strategy enhances classification accuracy. Specifically, using the Adam optimizer, the proposed method achieved a classification accuracy of 98.71% on the Plant Pathology database, representing an approximately 3% improvement and outperforming state-of-the-art methods. This demonstrates the effectiveness of background removal as a data augmentation technique for improving the robustness of disease classification models in real-world conditions
"Enfer et folie", propositions pour une lecture d'un texte de Youcef Sebti
Hell and madness, an analysis of a poem by Youcef Sebti. Hell and madness, a short poem published in Anthologie de la nouvelle poésie algérienne, edited by Jean Sénac, is analysed without reference either to the author or to his other works. However, his situation as a citizen of a country now independent after a hundred years of colonization is not ignored. This is an attempt to uncover the deeper meaning of the poem, by dint of sylistic analysis. The problem — to which no definite solution is found — is the following : which value, positive or negative, can be accorded to the second pole of the poem, namely madness ? It seems possible to interpret the whole poem as the life-journey of a man, first through the hell of colonization (or in a more general sense, of oppression) followed by the struggle for independence. But even when he thinks he is out of it, he has to admit that the hell still exists, and that the only way out of it lies through madness. The fact that madness is metaphorically linked with the sea may, along with other indications, show that madness is a kind of liberation, a flight to heaven.L 'enfer et la folie, un court poème paru dans l'anthologie de la nouvelle poésie algérienne de Jean Sénac, est analysé sans référence à l'auteur, ni au reste de son œuvre. Toutefois il est tenu compte de la situation : c'est un "décolonisé" qui a écrit ce poème. Une approche stylistique, serrant le texte de très près, essaie d'en dévoiler la signification profonde. Le principal problème qui se pose, et qui n'est pas résolu définitivement, est celui de la valeur positive ou négative qui doit être accordée aux deuxième pôle du texte : la folie. Il semble qu'on puisse interpréter l'ensemble comme l'itinéraire d'un être humain qui connaît d'abord l'enfer de l'oppression et qui se croit libéré, mais il comprend ensuite que l'enfer subsiste et que la seule issue est dans la folie. Cette folie, liée métaphoriquement à la mer, peut être considérée, semble-t-il, comme une libération, une évasion vers un paradis inconnu.Seferian Marie Alice. "Enfer et folie", propositions pour une lecture d'un texte de Youcef Sebti. In: Revue de l'Occident musulman et de la Méditerranée, n°22, 1976. pp. 181-187
L’approche pragmatique dans le discours théâtral pour enfants. Cas de la pièce "Les ailes de Namoula" de Youcef Baaloudj
The main aim of this study is to question the process of putting into theatrical discourse the perspective opened by the pragmatics and enunciative theory. We propose a reflection on the functioning of discursive mechanisms and the construction of the textual world of theatrical genre intended for children through the identification of linguistic and non-linguistic means provided by the author to guide the interpretation. Also through the different discursive mechanisms that accompany and manage the whole operation of putting into discourse namely the implicit, intertextuality, dialogism and the functioning of illocutory acts to deduce the morality underlying the children’s play : the wings of Namoula ""أجنحة نمولة by Youcef Baaloudj. Results show that the accentuation of the use of illocutory acts by the characters of the piece as well as the different forms of intertextuality. This is in the songs repeated by the characters and the implicit incorporation of the tale of The Cicada and the Ant then the explicit integration of the tale The Rabbit and the Turtle.Cet article vise à réaliser une lecture pragmatique énonciative de la pièce théâtrale destinée à l’enfant algérian أجنحة نمولة' ' ‘Les ailes de Namoula’ de Youcef Baaloudj. Notre démarche consiste à repérer les moyens linguistiques et non linguistiques utilisés par l’auteur pour tracer le chemin interprétatif de son public et délimiter les mécanismes discursifs qui fonctionnent au sein de ce discours .L’analyse révèle un emploi remarquable des actes illocutoires par les personnages, cet emploi fonctionne comme un trope communicationnel où l’énonciataire extrascénique (l’enfant) est impliqué indirectement. La présence de l’intertextualité est également révélatrice ; outre la fonction ludique des chansons et des fables incorporées : la morale tirée de La Cigale et la Fourmi ainsi que celle du Lièvre et la Tortue interpelle l’enfant pour créer des passerelles qui guident son interprétation de la présente pièce
Electrostatic force regularization for neural structured pruning
International audienceThe demand for deploying deep convolutional neural networks (DCNNs) on resource-constrained devices for real-time applications remains substantial. However, existing state-of-the-art structured pruning methods often involve intricate implementations, require modifications to the original network architectures, and necessitate an extensive fine-tuning phase. To address these challenges, we introduce a novel approach that integrates the concept of electrostatic force from physics into the training process of DCNNs. The magnitude of this force is directly proportional to the product of the charges of the convolution filter and the source filter, and inversely proportional to the square of the distance between them. We applied the electrostatic-like force to the convolution filters, attracting those with opposite charges toward non-zero weights and repelling similar ones toward zero. Consequently, repelled filters are pruned as their weights approach zero, while attracted filters retain significant parameters that preserve essential information. Unlike conventional methods, our approach is straightforward to implement, does not require any architectural modifications, and simultaneously optimizes weights and ranks filter importance, all without the need for extensive fine-tuning. We validated the efficacy of our method on modern DCNN architectures using the MNIST, CIFAR, and ImageNet datasets, achieving competitive performance compared to existing structured pruning approaches
