Predicting urban traffic is of great importance to smart city systems and public security; however, it is a very challenging task because of several dynamic and complex factors, such as patterns of urban geographical location, weather, seasons, and holidays. To tackle these challenges, we are stimulated by the deep-learning method proposed to unlock the power of knowledge from urban computing and proposed a deep-learning model based on neural network, entitled Capsules TCN Network, to predict the traffic flow in local areas of the city at once. Capsules TCN Network employs a Capsules Network and Temporal Convolutional Network as the basic unit to learn the spatial dependence, time dependence, and external factors of traffic flow prediction. In specific, we consider some particular scenarios that require accurate traffic flow prediction (e.g., smart transportation, business circle analysis, and traffic flow assessment) and propose a GAN-based superresolution reconstruction model. Extensive experiments were conducted based on real-world datasets to demonstrate the superiority of Capsules TCN Network beyond several state-of-the-art methods. Compared with HA, ARIMA, RNN, and LSTM classic methods, respectively, the method proposed in the paper achieved better results in the experimental verification.

Using the intelligent image recognition method to recognize the reading of the unintelligent manual reading meter is an important approach to expand the sensing means and reduce the cost of transformation of the Industrial Internet of Things. Traditional methods based on morphology and back propagation neural networks are difficult to solve the problems of shadow, deflection, and dimness in the industrial environment. In this article, the deep learning‐based capsule networks for pointer instruments and digital display instruments were proposed to implement unintelligent manual reading meter recognition using. Due to the complex structure and a large number of parameters, the basic capsule network is difficult to run on devices with low computing resources. To overcome this issue, the improved capsule network model was designed to reduce the parameters of the model. The experimental results show that the improved capsule network improves the recognition accuracy by 13% and reduces the parameters by 23%.