Search

Fully supervised deep learning (DL) methods have been widely used in low-dose CT (LDCT) imaging field and can usually achieve high accuracy results. These methods require a large labeled training set which consists of pairs of LDCT images as well as their corresponding high-dose CT (HDCT) ones. They successfully learn intermediate concept of features describing important components in CT images, such as noise distribution, and structure details, which is important to capture dependencies from LDCT image to HDCT ones. However, it should be noted that it is quite time-consuming and costly to obtain such a large of labeled CT images especially the HDCT images are limited in clinics. In comparison, lots of unlabeled LDCT images are usually easily accessible and massive critical information latent in the unlabeled LDCT can be leveraged to further boost restoration performance. Therefore, in this work, we present a semi-supervised noise distribution learning network to suppress noise-induced artifacts in the LDCT images. For simplicity, the presented network in termed as "SNDL-Net". The presented SNDL-Net consists of two sub-networks, i.e., supervised network, and unsupervised network. In the supervised network, the LDCT/HDCT image pairs are used for network training. And the unsupervised network considers the complex noise distribution in the LDCT images, and model the noise with a Gaussian mixture framework, then learns the proper gradient of LDCT images in a purely unsupervised manner. Similar with the supervised network training, the gradient information in a large of unlabeled LDCT images can be used for unsupervised network training. Moreover, to learn the noise distribution accurately, the discrepancy between the learned noise distribution in the supervised network and learned noise distribution in the unsupervised network can be modeled by a Kullback-Leibler (KL) divergence. Experiments on the Mayo clinic dataset verify the method is effective in low-dose CT image restoration with only a small amount of labeled data compared to previous supervised deep learning methods.

A new type of information processing system is developed. The processing is implemented by usinghybrid architecture of VanderLugt optical correlator and joint transform correlator. In one opticalsetup, joint transform correlator is used for object tracking and Vander-Lugt optical correlator isused for object recognition respectively. The experiments show that we can get good results.Key words: optical correlator, object tracking, pattern recognition 1. INTRODUCTION There are mainly two approaches in optical correlator; namely, the Vander-Lugt correlator(VLC)and the joint transform correlator(JTC). VLC is simple and has high signal-to-noise ratio. However, the fabrication of the matched filter is inconvenient. VLC is not suitable for real-time object recognition, so its application range is restricted. During recent years, JTC has been well developed.Some useflul arithmetics and architectures have been introduced. The major advantage ofJTC is that synthesis of a complex spatial filter like a matched filter is not necessary to execute matched filtering.

A multichannel Vander-Lugt correlator based on Ferroelectric Liquid Crystal Spatial Light Modulator is proposed in this paper. It takes advantage of the existence of higher diffraction orders generated by the pixeling effect of electrically addressed SLM (EASLM). The preliminary experimental results are presented.© (1996) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A miniature optical pattern recognition (OPR) system is proposed in this paper. We use a multiple channel optical correlator to perform real-time image processing, such as dynamic multiple objects recognition, nonlinear operation of fusion of correlation output, and compression and reconstruction of infrared fingerprint images. The preliminary experimental results show that a two-dimensional optical correlator based on binary optics makes the OPR system to tend to practice.

A micro optics of 24 X 24 binary phase-only Fresnel lens array with eight discrete phase levels has been made on a quartz substrate. The diffraction efficiency is about 80%. Being of the functions of beamsplitter and Fourier transformation, it can be used in multiple optical correlator to enlarge the memory capacity for target recognition and to make the system smaller, lighter, inexpensive and good accuracy.