Support vector machines versus multi-layer perceptrons for efficient off-line signature recognition

Frı́as-Martı́nez, Enrique; Sánchez, Ángel Mediavilla; Vélez, José Francisco

doi:10.1016/j.engappai.2005.12.006

Cited by 109 publications

(50 citation statements)

References 32 publications

(56 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Por outro lado, no MLP (multilayer perceptron) a minimização do erro nos dados de treino é feita da forma empírica tradicional. A principal diferença na minimização do risco é melhorar o desempenho da generalização do SVM em comparação a um MLP [6].…”

Section: Suppot Vector Machine (Svm)unclassified

“…A complexidade algorítmica da busca em uma kd-tree implica em um tempo de execução cada vez maior em função da quantidade de dados de treino na técnica k-vizinhos mais próximos. Para as técnicas SVM e RNA, onde a construção do hiperplano de separação tem seu custo apenas na inicialização [6], identifica-se que possuem um tempo de busca praticamente constante independente da quantidade de pontos. Nos resultados foi apresentada a possibilidade de inclusão de novos grupos de pontos na estatística de teste, essa ação poderia levar a uma conclusão mais definitiva, mostrando perfeitamente o momento da experimentação de um máximo no comportamento entre desvio padrão/número de pontos de teste (Tabelas 1 a 6) e ainda expor que a técnica SVM continuará com o seu comportamento estável.…”

Section: Figuraunclassified

See 1 more Smart Citation

Avaliação do desempenho de classificadores em metalografia quantitativa

Sarruf¹,

Cidade²,

Braga³

et al. 2015

Matéria (Rio J.)

View full text Add to dashboard Cite

RESUMOAs técnicas convencionais para aplicação da metalografia quantitativa têm sido cada vez mais substituídas por processos digitais devido à crescente necessidade de ganho de velocidade nas análises industriais e o processamento de imagens tem então ocupado um lugar importante nesse campo da metalurgia.Este trabalho teve como objetivo a implementação e avaliação de técnicas de inteligência e visão computacional na resolução de problemas de determinação de área de fases em aços AISI 1020. Para a comparação, três técnicas de decisão automática foram avaliadas: k-vizinhos mais próximos (KNN), que classifica os pontos da imagem através de correlação com a sua vizinhança e duas técnicas de classificação por aprendizado supervisionado: Redes Neurais Artificiais (RNA) e Máquinas de Vetores Suporte (SVM). Índices de indicação de qualidade de classificação também foram avaliados.Os resultados mostraram superioridade geral na técnica SVM no que tange, principalmente, o tempo de classificação e acurácia relativa, porém em todos os casos houve concordância com os valores de fração de área esperados para esse tipo de material, segundo análises metalúrgicas teóricas. Palavras-chave:Classificadores de Imagem, Metalografia quantitativa, Processamento de imagens, Caracterização microestrutural. ABSTRACTAs the need for increasing speed takes place at industrial processes in general, the use of digital techniques such as image processing and automatic classification have been playing an important role at materials characterization and quantitative metallography fields.The aim of this work was to develop and evaluate computational vision techniques on solving ordinary problems as the area fraction of phases determination of AISI 1020 steels. Three techniques were implemented and evaluated: k-Nearest Neighbors (KNN) which consists in classifying pixels based on their neighborhood information, Artificial Neural Networks and Support Vectors Machine, these last two centered on supervised machine learning processes. Indexes that denote classification quality were then evaluated.Concerning classification time and relative accuracy, the SVM results have shown superiority. Nevertheless, in all cases, the classification values have agreed with the area fraction values expected for this type of steel based on theoretical metallurgical analysis.

show abstract

Section: Suppot Vector Machine (Svm)unclassified

Section: Figuraunclassified

Avaliação do desempenho de classificadores em metalografia quantitativa

Sarruf¹,

Cidade²,

Braga³

et al. 2015

Matéria (Rio J.)

View full text Add to dashboard Cite

show abstract

“…The functionality of the topology of the MLP is determined by a learning algorithm able to modify the parameters of the net. The algorithm of back propagation, based on the method of steepest descent [4] in the process of upgrading the connection weights, is the most commonly used by the scientific community. Multi-Layer Perceptrons (MLP) are fully connected feed forward nets with one or more layers of nodes between the input and the output nodes.…”

Section: Brief Literature Surveymentioning

confidence: 99%

Pattern Recognition System using MLP Neural Networks

Chauhan¹

2012

IOSRJEN

View full text Add to dashboard Cite

Pattern recognition can be used to recognize and/or locate specific objects in an image. The pattern recognition approaches are based on analysis of statistical parameters computed using image processing tools. The parameters may be compared with the standard pattern parameters to identify the pattern or a neural network may be trained using the statistical parameters to identify a given pattern. In the presented work, a neural network approach has been worked out in identifying a pattern. The neural approach applies biological concepts to machines to recognize patterns. A neural network is an information processing system. It consists of massive simple processing units with a high degree of interconnection between each unit. The processing units work cooperatively with each other and achieve massive parallel distributed processing. The design and function of neural networks simulate some functionality of biological brains and neural systems. The advantages of neural networks are their adaptive-learning, self-organization and fault-tolerance capabilities. For these outstanding capabilities, neural networks are used for pattern recognition applications. In the presented work, multi-layer perceptron (MLP) neural network is used for pattern recognition. Here, each neuron computes a so called "net input" from the outputs of previous neurons and from the weights of the connections. Typically, such a net input is a weighted sum, and a numerical value, called "bias", is added to the net input. In MLPs, a function called "activation function", is applied to the net input. In our experiments, we used a sigmoid function.

show abstract

“…Some techniques employed in the area of signature recognition are: application of vector quantization -dynamic time warping scheme for signature recognition [13], applications of active deformable models for approximating the external shape of the signatures [14], [15] comparison of support vector machines and Neural Network for signature recognition is made in [15], Comparison of support vector machines and multilayer perceptrons for signature recognition [16].…”

Section: Introductionmentioning

confidence: 99%

Retrieval of Offline Handwritten Signatures

Prakash¹,

Guru²

2010

IJCA

View full text Add to dashboard Cite

Similarity retrieval of images is an important task in database applications. In such applications, effective organization and retrieval of images can be achieved through indexing. In this paper, the problem of quick retrieval of offline signatures in the context of database of signature images is addressed. The proposed methodology retrieves signatures in the database of signature images for a given query signature according to the decreasing order of their spatial similarity with the query. Similarity computed is based on orientations of corresponding edges drawn in between geometric centers (centroids) of the signature image. We retrieve the best hypotheses in a simple yet efficient way to speed up the subsequent robust recognition stage. The runtime of the signature recognition process is reduced, because the scanning of the entire database for a given query is narrowed down to comparing the query with a few top retrieved hypotheses. The experimentation conducted on a large MCYT_signature database [1] has shown promising results. The results demonstrate the efficacy of the proposed methodology.

show abstract

Support vector machines versus multi-layer perceptrons for efficient off-line signature recognition

Cited by 109 publications

References 32 publications

Avaliação do desempenho de classificadores em metalografia quantitativa

Avaliação do desempenho de classificadores em metalografia quantitativa

Pattern Recognition System using MLP Neural Networks

Retrieval of Offline Handwritten Signatures

Contact Info

Product

Resources

About