Matches in SemOpenAlex for { <https://semopenalex.org/work/W4322751519> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4322751519 abstract "Fingerprint images from crime scenes can be used to find and identify suspects in the field of forensic science. Fingerprint images are usually polluted by noise, which affects the visual effect of fingerprint images. It is important to process the noise of fingerprint images. A supervised learning 9-layer artificial neural network besides the input layer and the output layer is designed to remove the noise of fingerprint images in this paper. The first three layers of the neural network are convolutional layers. The purpose of designing convolutional layers is to extract the feature information of the images layer by layer and gradually generate the feature maps. The fourth layer to the sixth layer are fully connected layers. The number of neurons in the fourth layer is the same as that of the sixth layer. The number of neurons in the fourth layer is the number of elements in the vector flattened by the feature maps outputted from the last convolutional layer. The seventh layer to the ninth layer in the neural network are deconvolutional layers. These deconvolutional layers gradually restore the feature maps to the fingerprint image. Finally, the enhanced fingerprint image is outputted from the output layer. As the training samples of fingerprint images is inadequate, this paper provides a way to make training samples. The clear fingerprint image is blurred as the corresponding blurred fingerprint image, which solves the problem of lack of the training samples. The blurred fingerprint image and the corresponding clear fingerprint image are used to establish the training sample set and train the artificial neural network. The mean square error function is used as the loss function to adjust and optimize the parameters of the neural network. The neural network in this paper is evaluated on test fingerprint images. The neural network method is compared with the Gaussian low-pass filtering method and the Wiener filtering method. Experimental results show that the neural network method is superior to the Gaussian low-pass filtering method and the Wiener filtering method in filtering Gaussian noise of fingerprint images." @default.
- W4322751519 created "2023-03-03" @default.
- W4322751519 creator A5091062044 @default.
- W4322751519 date "2022-07-01" @default.
- W4322751519 modified "2023-10-16" @default.
- W4322751519 title "Artificial Neural Network for Processing Fingerprint Image Noise" @default.
- W4322751519 cites W1966954983 @default.
- W4322751519 cites W2074752705 @default.
- W4322751519 cites W2122249759 @default.
- W4322751519 cites W2611611254 @default.
- W4322751519 cites W2800017313 @default.
- W4322751519 cites W2891006103 @default.
- W4322751519 cites W2893483035 @default.
- W4322751519 cites W2911075534 @default.
- W4322751519 cites W2962700793 @default.
- W4322751519 cites W3102737931 @default.
- W4322751519 cites W3106105822 @default.
- W4322751519 doi "https://doi.org/10.1109/snpd-summer57817.2022.00011" @default.
- W4322751519 hasPublicationYear "2022" @default.
- W4322751519 type Work @default.
- W4322751519 citedByCount "0" @default.
- W4322751519 crossrefType "proceedings-article" @default.
- W4322751519 hasAuthorship W4322751519A5091062044 @default.
- W4322751519 hasConcept C115961682 @default.
- W4322751519 hasConcept C138885662 @default.
- W4322751519 hasConcept C153180895 @default.
- W4322751519 hasConcept C154945302 @default.
- W4322751519 hasConcept C159985019 @default.
- W4322751519 hasConcept C168406668 @default.
- W4322751519 hasConcept C192562407 @default.
- W4322751519 hasConcept C2776401178 @default.
- W4322751519 hasConcept C2777826928 @default.
- W4322751519 hasConcept C2779227376 @default.
- W4322751519 hasConcept C31972630 @default.
- W4322751519 hasConcept C41008148 @default.
- W4322751519 hasConcept C41895202 @default.
- W4322751519 hasConcept C50644808 @default.
- W4322751519 hasConcept C52622490 @default.
- W4322751519 hasConcept C81363708 @default.
- W4322751519 hasConcept C99498987 @default.
- W4322751519 hasConceptScore W4322751519C115961682 @default.
- W4322751519 hasConceptScore W4322751519C138885662 @default.
- W4322751519 hasConceptScore W4322751519C153180895 @default.
- W4322751519 hasConceptScore W4322751519C154945302 @default.
- W4322751519 hasConceptScore W4322751519C159985019 @default.
- W4322751519 hasConceptScore W4322751519C168406668 @default.
- W4322751519 hasConceptScore W4322751519C192562407 @default.
- W4322751519 hasConceptScore W4322751519C2776401178 @default.
- W4322751519 hasConceptScore W4322751519C2777826928 @default.
- W4322751519 hasConceptScore W4322751519C2779227376 @default.
- W4322751519 hasConceptScore W4322751519C31972630 @default.
- W4322751519 hasConceptScore W4322751519C41008148 @default.
- W4322751519 hasConceptScore W4322751519C41895202 @default.
- W4322751519 hasConceptScore W4322751519C50644808 @default.
- W4322751519 hasConceptScore W4322751519C52622490 @default.
- W4322751519 hasConceptScore W4322751519C81363708 @default.
- W4322751519 hasConceptScore W4322751519C99498987 @default.
- W4322751519 hasLocation W43227515191 @default.
- W4322751519 hasOpenAccess W4322751519 @default.
- W4322751519 hasPrimaryLocation W43227515191 @default.
- W4322751519 hasRelatedWork W1502614025 @default.
- W4322751519 hasRelatedWork W1925241029 @default.
- W4322751519 hasRelatedWork W2059299633 @default.
- W4322751519 hasRelatedWork W2067275498 @default.
- W4322751519 hasRelatedWork W2732542196 @default.
- W4322751519 hasRelatedWork W2734991885 @default.
- W4322751519 hasRelatedWork W2760085659 @default.
- W4322751519 hasRelatedWork W2940977206 @default.
- W4322751519 hasRelatedWork W3193765978 @default.
- W4322751519 hasRelatedWork W4303683349 @default.
- W4322751519 isParatext "false" @default.
- W4322751519 isRetracted "false" @default.
- W4322751519 workType "article" @default.