Satellite Image Processing Systems: An Architectural Perspective

Satellite image processing systems include satellite image classification, long ranged data processing, yield prediction systems, etc. All of these systems require a large quantity of images for effective processing, and thus they are directed towards big-data applications. All these applications require a series of highly complex image processing and signal processing steps, which include but are not limited to image acquisition, image pre-processing, segmentation, feature extraction & selection, classification and post processing. Numerous researchers globally have proposed a large variety of algorithms, protocols and techniques in order to effectively process satellite images. This makes it very difficult for any satellite image system designer to develop a highly effective and application-oriented processing system. In this paper, we aim to categorize these large number of researches w.r.t. their effectiveness and further perform statistical analysis on the same. This study will assist researchers in selecting the best and most optimally performing algorithmic combinations in order to design a highly accurate satellite image processing system.

Light-induced perovskite dynamic transformation enabling photodetector mimicking neuromorphic vision sensing system

As an artificial perception system, neuromorphic vision sensing system can imitate the complex image sensing and processing functions of the human visual neural network. In order to stimulate the nervous...

Semantic Segmentation of Remote Sensing Image Based on GAN and FCN Network Model

Accurate remote sensing image segmentation can guide human activities well, but current image semantic segmentation methods cannot meet the high-precision semantic recognition requirements of complex images. In order to further improve the accuracy of remote sensing image semantic segmentation, this paper proposes a new image semantic segmentation method based on Generative Adversarial Network (GAN) and Fully Convolutional Neural Network (FCN). This method constructs a deep semantic segmentation network based on FCN, which can enhance the receptive field of the model. GAN is integrated into FCN semantic segmentation network to synthesize the global image feature information and then accurately segment the complex remote sensing image. Through experiments on a variety of datasets, it can be seen that the proposed method can meet the high-efficiency requirements of complex image semantic segmentation and has good semantic segmentation capabilities.

Cloud-enabled Fiji for reproducible, integrated and modular image processing

Biomedical image analyses can require many steps processing different types of data. Analysis of increasingly large data sets often exceeds the capacity of local computational resources. We present an easy-to-use and modular cloud platform that allows biomedical researchers to reproducibly execute and share complex analytical workflows to process large image datasets. The workflows and the platform are encapsulated in software containers to ensure reproducibility and facilitate installation of even the most complicated workflows. The platform is both graphical and interactive allowing users to use the viewer of their choice to adjust the image pre-processing and analysis steps to iteratively improve the final results. We demonstrate the utility of our platform via two use cases in focal adhesion and 3D imaging analyses. In particular, our focal adhesion workflow demonstrates integration of Fiji with Jupyter Notebooks. Our 3D imaging use case applies Fiji/BigStitcher to big datasets on the cloud. The accessibility and modularity of the cloud platform democratizes the application and development of complex image analysis workflows.

Selection of Key Frames for 3D Reconstruction in Real Time

Cameras play a prominent role in the context of 3D data, as they can be designed to be very cheap and small and can therefore be used in many 3D reconstruction systems. Typical cameras capture video at 20 to 60 frames per second, resulting in a high number of frames to select from for 3D reconstruction. Many frames are unsuited for reconstruction as they suffer from motion blur or show too little variation compared to other frames. The camera used within this work has built-in inertial sensors. What if one could use the built-in inertial sensors to select a set of key frames well-suited for 3D reconstruction, free from motion blur and redundancy, in real time? A random forest classifier (RF) is trained by inertial data to determine frames without motion blur and to reduce redundancy. Frames are analyzed by the fast Fourier transformation and Lucas–Kanade method to detect motion blur and moving features in frames to label those correctly to train the RF. We achieve a classifier that omits successfully redundant frames and preserves frames with the required quality but exhibits an unsatisfied performance with respect to ideal frames. A 3D reconstruction by Meshroom shows a better result with selected key frames by the classifier. By extracting frames from video, one can comfortably scan objects and scenes without taking single pictures. Our proposed method automatically extracts the best frames in real time without using complex image-processing algorithms.

Current approaches to fate mapping and lineage tracing using image data

ABSTRACT Visualizing, tracking and reconstructing cell lineages in developing embryos has been an ongoing effort for well over a century. Recent advances in light microscopy, labelling strategies and computational methods to analyse complex image datasets have enabled detailed investigations into the fates of cells. Combined with powerful new advances in genomics and single-cell transcriptomics, the field of developmental biology is able to describe the formation of the embryo like never before. In this Review, we discuss some of the different strategies and applications to lineage tracing in live-imaging data and outline software methodologies that can be applied to various cell-tracking challenges.

Deep learning for bioimage analysis in developmental biology

ABSTRACT Deep learning has transformed the way large and complex image datasets can be processed, reshaping what is possible in bioimage analysis. As the complexity and size of bioimage data continues to grow, this new analysis paradigm is becoming increasingly ubiquitous. In this Review, we begin by introducing the concepts needed for beginners to understand deep learning. We then review how deep learning has impacted bioimage analysis and explore the open-source resources available to integrate it into a research project. Finally, we discuss the future of deep learning applied to cell and developmental biology. We analyze how state-of-the-art methodologies have the potential to transform our understanding of biological systems through new image-based analysis and modelling that integrate multimodal inputs in space and time.

Recent trends in AFM imaging speed and improvement methods

: Atomic Force Microscope (AFM) has become the primary tool for observation and manipulation in nanotechnology research due to its nano-meter high resolution. However, the slow imaging speed is one of the critical reasons hindering the further development of AFM. This article first introduces the applications of AFM in cell biology in recent years, then expresses the importance of rapid imaging in cell biology, and finally summarizes the reasons affecting the imaging speed of AFM from three aspects: the limited bandwidth of system mechanical components, obvious inherent characteristics of piezoelectric scanners, and complex image processing algorithms. The improvement and optimization methods of mechanical parts or structure, control algorithm, and image processing are reviewed for different influence reasons. Then, the advantages of different improvement methods and improved imaging speed are discussed, and imaging quality improvement effects are compared. Imaging speed and resolution both are much higher than before while ensuring image quality without damaging the samples. This review aims to enable students, the public, and even experts of different knowledge backgrounds to learn directly and select realizable improvement method according to realistic conditions. Finally, the future development trend and further prospects of high-speed AFM are discussed.