Image demosaicking has been an essential and challenging problem among the most crucial steps of image processing behind image sensors. Due to the rapid development of intelligent processors based on deep learning, several demosaicking methods based on a convolutional neural network (CNN) have been proposed. However, it is difficult for their networks to run in real-time on edge computing devices with a large number of model parameters. This paper presents a compact demosaicking neural network based on the UNet++ structure. The network inserts densely connected layer blocks and adopts Gaussian smoothing layers instead of down-sampling operations before the backbone network. The densely connected blocks can extract mosaic image features efficiently by utilizing the correlation between feature maps. Furthermore, the block adopts depthwise separable convolutions to reduce the model parameters; the Gaussian smoothing layer can expand the receptive fields without down-sampling image size and discarding image information. The size constraints on the input and output images can also be relaxed, and the quality of demosaicked images is improved. Experiment results show that the proposed network can improve the running speed by 42% compared with the fastest CNN-based method and achieve comparable reconstruction quality as it on four mainstream datasets. Besides, when we carry out the inference processing on the demosaicked images on typical deep CNN networks, Mobilenet v1 and SSD, the accuracy can also achieve 85.83% (top 5) and 75.44% (mAP), which performs comparably to the existing methods. The proposed network has the highest computing efficiency and lowest parameter number through all methods, demonstrating that it is well suitable for applications on modern edge computing devices.
It is more and more common to use function words as an important text feature of Chinese, such as the research on “A Dream of Red Mansions” of Li Xianping. But the effect of using all function words as a feature in distinguishing writers' writing style is not prominent. Our study finds that using the classical Chinese function words and sentence tail function words as a feature is better than differentiated by the classification of all function words, and the experiments have proved this point. At the same time, with the attempts of weighting two function words features, this paper finds the most effective weighted coefficients, which can further improve the classification accuracy, establishes the mathematical model which can effectively judge the author's style of writing, and makes a comparative study of some major novels in the Ming and Qing dynasties by using SVM. And this paper also probes into the identity of the author of “The Plum In The Golden Vase”. At the same time, we proved that “San-Yan” and “The Plum In The Golden Vase” are not written by the same person, and “A Dream of Red Mansions” have a different writer for the late 40 chapters.