Andere Kunden interessierten sich auch für
![Cognitive Lightweight Management Systems in the Underwater-IoT]( "Cognitive Lightweight Management Systems in the Underwater-IoT")
Cognitive Lightweight Management Systems in the Underwater-IoT24,99 €![Machine Learning and its Applications]( "Machine Learning and its Applications")
Machine Learning and its Applications59,99 €![Underwater Acoustic Data Processing]( "Underwater Acoustic Data Processing")
Underwater Acoustic Data Processing231,99 €![Machine Learning Based Subtitles Detection With NLP]( "Machine Learning Based Subtitles Detection With NLP")
Machine Learning Based Subtitles Detection With NLP26,99 €![Prediction of breast cancer using various machine learning methods]( "Prediction of breast cancer using various machine learning methods")
Prediction of breast cancer using various machine learning methods36,99 €![Machine Learning Algorithms Using Classical and Quantum Photonics]( "Machine Learning Algorithms Using Classical and Quantum Photonics")
Machine Learning Algorithms Using Classical and Quantum Photonics50,99 €![Regression Testing in Era of Internet of Things and Machine Learning]( "Regression Testing in Era of Internet of Things and Machine Learning")
Regression Testing in Era of Internet of Things and Machine Learning50,99 €
Image performance in underwater robots is one of the most challenging problems for autonomous underwater robotics due to light transmission in water. Although image restoration techniques can effectively remove a haze from a damaged image, they require multiple images from the same location making it difficult to use in real time. Considering the positive effects of in-depth learning strategies on other image processing problems such as coloring or finding objects, a deeper learning solution is proposed. The convolutional neural network is trained in image retrieval techniques to capture one image better than other image enhancement techniques. The proposed method is capable of producing high quality image restoration images with a single image as input. The neural network is verified using images from various locations and signals to prove the power of normal action.