Significance: Researchers have made great progress in single-image super-resolution (SISR) using deep convolutional neural networks. However, in the field of leukocyte imaging, the performance of existing SISR methods is still limited as it fails to thoroughly explore the geometry and structural consistency of leukocytes. The inaccurate super-resolution (SR) results will hinder the pathological study of leukocytes, since the structure and cell lineage determine the types of leukocyte and will significantly affect the subsequent inspection.
Aim: We propose a deep network that takes full use of the geometry prior and structural consistency of the leukocyte images. We establish and annotate a leukocyte dataset, which contains five main types of leukocytes (basophil, eosinophil, monocyte, lymphocyte, and neutrophil), for learning the structure and geometry information.
Approach: Our model is composed of two modules: prior network and SR network. The prior network estimates the parsing map of the low-resolution (LR) image, and then the SR network takes both the estimated parsing map and LR image as input to predict the final high-resolution image.
Result: Experiments show that the geometry prior and structural consistency in use obviously improves the SR performance of leukocyte images, enhancing the peak-signal-to-noise ratio (PSNR) by about 0.4 dB in our benchmark.
Conclusion: As proved by our leukocyte SR benchmark, the proposed method significantly outperforms state-of-the-art SR methods. Our method not only improves the PSNR and structural similarity indices, but also accurately preserves the structural details of leukocytes. The proposed method is believed to have potential use in the wide-field cell prescreening by simply using a low-power objective.
Both light-field and polarization information contain lots of clues about scenes, and they can be widely used in variety of computer vision tasks. However, existing imaging systems cannot simultaneously capture the light-field and polarization information. In this paper, we present a low-cost and high-performance miniaturized polarimetric light-field camera, which is based on the six heterogeneous sensors array. The main challenge for the proposed strategy is to align the multi-view images with different polarization characteristics, especially for regions with high degree of polarization -- in which the intensity correlations are commonly weak. To solve this problem, we propose to use Convolutional Neural Network (CNN) based stereo matching method for aligning the heterogeneously polarized images accurately. After stereo matching, both the light field and the Stokes vectors of scene are estimated, and the polarimetry conventions, e.g., the polarization angle, the linear polarization degree and the circular polarization degree, are given. We implement the prototype of the multisensor polarization light-field camera and perform extensive experiments on it. The polarimetric light-field camera achieves six live streaming on time and the heterogeneous processor of NVIDIA Jetson TX2 is exploited for image processing. Benefiting from the multi-sensor parallel polarization imaging and efficient parallel processing, the proposed system achieves promising performance on time resolution, signal-to-noise ratio. Besides, we develop the object recognition applications to show the superiorities of proposed system.
Proceedings Volume Editor (2)
This will count as one of your downloads.
You will have access to both the presentation and article (if available).
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.