Automated Detection, Classification and Counting of Fish in Fish Passages With Deep Learning

The Ocean Aware project, led by Innovasea and funded through Canada's Ocean Supercluster, is developing a fish passage observation platform to monitor fish without the use of traditional tags. This will provide an alternative to standard tracking technology, such as acoustic telemetry fish tracking, which are often not appropriate for tracking at-risk fish species protected by legislation. Rather, the observation platform uses a combination of sensors including acoustic devices, visual and active sonar, and optical cameras. This will enable more in-depth scientific research and better support regulatory monitoring of at-risk fish species in fish passages or marine energy sites. Analysis of this data will require a robust and accurate method to automatically detect fish, count fish, and classify them by species in real-time using both sonar and optical cameras. To meet this need, we developed and tested an automated real-time deep learning framework combining state of the art convolutional neural networks and Kalman filters. First, we showed that an adaptation of the widely used YOLO machine learning model can accurately detect and classify eight species of fish from a public high resolution DIDSON imaging sonar dataset captured from the Ocqueoc River in Michigan, USA. Although there has been extensive research in the literature identifying particular fish such as eel vs. non-eel and seal vs. fish, to our knowledge this is the first successful application of deep learning for classifying multiple fish species with high resolution imaging sonar. Second, we integrated the Norfair object tracking framework to track and count fish using a public video dataset captured by optical cameras from the Wells Dam fish ladder on the Columbia River in Washington State, USA. Our results demonstrate that deep learning models can indeed be used to detect, classify species, and track fish using both high resolution imaging sonar and underwater video from a fish ladder. This work is a first step toward developing a fully implemented system which can accurately detect, classify and generate insights about fish in a wide variety of fish passage environments and conditions with data collected from multiple types of sensors.

SDR-Implemented Passive Bistatic SAR System Using Sentinel-1 Signal and Its Experiment Results

A fixed-receiver mobile-transmitter passive bistatic synthetic aperture radar (MF-PB-SAR) system, which uses the Sentinel-1 SAR satellite as its non-cooperative emitting source, has been developed by using embedded software-defined radio (SDR) hardware for high-resolution imaging of the targets in a local area in this study. Firstly, Sentinel-1 and the designed system are introduced. Then, signal model, signal pre-processing methods, and effective target imaging methods are presented. At last, various experiment results of target imaging obtained at different locations are shown to validate the developed system and the proposed methods. It was found that targets in a range of several kilometers can be well imaged.

A case for glycerol as an acceptable additive for single-particle cryoEM samples

Buffer-composition and sample-preparation guidelines for cryo-electron microscopy are geared towards maximizing imaging contrast and reducing electron-beam-induced motion. These pursuits often involve the minimization or the complete removal of additives that are commonly used to facilitate proper protein folding and minimize aggregation. Among these admonished additives is glycerol, a widely used osmolyte that aids protein stability. In this work, it is shown that the inclusion of glycerol does not preclude high-resolution structure determination by cryoEM, as demonstrated by an ∼2.3 Å resolution reconstruction of mouse apoferritin (∼500 kDa) and an ∼3.3 Å resolution reconstruction of rabbit muscle aldolase (∼160 kDa) in the presence of 20%(v/v) glycerol. While it was found that generating thin ice that is amenable to high-resolution imaging requires long blot times, the addition of glycerol did not result in increased beam-induced motion or an inability to pick particles. Overall, these findings indicate that glycerol should not be discounted as a cryoEM sample-buffer additive, particularly for large, fragile complexes that are prone to disassembly or aggregation upon its removal.

Progress in design, nanofabrication and performance of metalenses

Metalens has been shown to overcome the diffraction limit of conventional optical lenses to achieve sub-wavelength resolution. Due to its planar structure and lightweight, metalens has the potential applications in the manufacture of flat lenses for cameras and other high resolution imaging optics. However, currently reported metalenses have low focusing efficiencies: 26% - 68% in THz and GHz range, 1% - 91% in near infrared range (NIR), and 5% - 91.6% in the visible range. Far field imaging in the visible light is essential for use in camera and mobile phones, which requires a complex metalens structure with multi-layers of alternating metal and dielectric layers. Most of the reported metalenses work in a single wavelength, mainly due to the high dispersion characteristics of the diffractive metalenses. It remains a challenge to realize high resolution imaging for a wide wavelength band in particular in the visible range. In this review, we report the state-of-the-art in metalens design principle, types of nanoscale structures, and various fabrication processes. We introduce femtosecond laser direct writing based on two-photon polymerization as an emerging nanofabrication technology. We provide an overview of the optical performance of the recently-reported metalenses and elaborate the major research and engineering challenges and future prospects.

Geomorphic contexts and science focus of the Zhurong landing site on Mars

As part of the Tianwen-1 mission, the Zhurong rover successfully touched down in southern Utopia Planitia on 15 May 2021. On the basis of the new sub-metre-resolution images from the High Resolution Imaging Camera on board the Tianwen-1 orbiter, we determined that the Zhurong rover landed at 109.925° E, 25.066° N at an elevation of −4,099.4 m. The landing site is near the highland–lowland boundary1 and multiple suspected shorelines2–7. Under the guidance of the remote sensing survey, the Zhurong rover is travelling south for specific in situ investigation. Supported by the six payloads on board the rover8, its initial key targets are rocks, rocky fields, transverse aeolian ridges and subsurface structures along the path. Extended investigation will aim at troughs and cones in the distance. A better understanding of the formation mechanisms of these targets may shed light on the historical volcanism and water/ice activities within the landing area, as well as the activities of the wind. These results may reveal the characteristics and evolution of the ancient Martian environment and advance the exploration of the habitability of ancient Mars.