Analysis and Research of Image Processing Optimization Technology based on Big Data Processing

Convolutional neural network algorithm is a kind of deep neural network algorithm, which has good performance of image feature extraction. It can reduce the large data amount of images to a small data amount without affecting the feature expression, and effectively solve the problem of low image rec...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	2024 Second International Conference on Data Science and Information System (ICDSIS) s. 1 - 5
Hlavní autori:	Chen, Yelei, Lan, Ruixue
Médium:	Konferenčný príspevok..
Jazyk:	English
Vydavateľské údaje:	IEEE 17.05.2024
Predmet:	Accuracy Big Data big data processing technology Computational modeling convolutional neural network image processing optimization technology Image recognition Mean square error methods PSNR Training
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Popis
Shrnutí:	Convolutional neural network algorithm is a kind of deep neural network algorithm, which has good performance of image feature extraction. It can reduce the large data amount of images to a small data amount without affecting the feature expression, and effectively solve the problem of low image recognition accuracy. In this paper, the structure of convolutional neural network algorithm is studied deeply, and a 10-layer convolutional neural network model is designed according to the requirements of computer image recognition. Through image transformation and backpropagation algorithm training, the overfitting problem in model training is effectively controlled. The experimental results show that the algorithm has smaller peak signal-to-noise ratio, structural similarity coefficient and mean square error, and the running time is below 0.13s.
DOI:	10.1109/ICDSIS61070.2024.10594155