Recently, imaged-based approaches have developed rapidly for high-throughput plant phenotyping (HTPP). Imaging reduces a 3D into 2D images, which makes the retrieval of morphological traits challenging. We novel LiDAR-based instrument to generate point clouds single plants. The combined LiDAR scanner with precision rotation stage on an individual was placed. A LabVIEW program control scanning and motion, synchronize measurements from both devices, capture 360° view cloud. data processing...

With the increasing demand for search and rescue, it is highly demanded to detect objects of interest in large-scale images captured by Unmanned Aerial Vehicles (UAVs), which quite challenging due extremely small scales objects. Most existing methods employed Feature Pyramid Network (FPN) enrich shallow layers' features combing deep contextual features. However, under limitation inconsistency gradient computation across different layers, layers FPN are not fully exploited tiny In this paper,...

Summary A higher minimum (night‐time) temperature is considered a greater limiting factor for reduced rice yield than similar increase in maximum (daytime) temperature. While the physiological impact of high night (HNT) has been studied, genetic and molecular basis HNT stress response remains unexplored. We examined phenotypic variation mature grain size (length width) diverse set accessions under stress. Genome‐wide association analysis identified several HNT‐specific loci regulating as...

Summary Lignin is a key target for modifying lignocellulosic biomass efficient biofuel production. Brown midrib 12 ( bmr12 ) encodes the sorghum caffeic acid O‐methyltransferase (COMT) and one of enzymes in monolignol biosynthesis. Loss function mutations COMT reduces syringyl (S) lignin subunits improves conversion rate. Although plays an important role maintaining cell wall integrity xylem vessels, physiological molecular consequences due to loss on root growth adaptation water deficit...

Accurate measurement of seed size parameters is essential for both breeding efforts aimed at enhancing yields and basic research focused on discovering genetic components that regulate size. To address this need, we have developed an open-source graphical user interface (GUI) software, SeedExtractor determines shape (including area, perimeter, length, width, circularity, centroid), color with capability to process a large number images in time-efficient manner. In context, our application...

The use of 3D plant models for high-throughput phenotyping is increasingly becoming a preferred method many science researchers. Numerous camera-based imaging systems and reconstruction algorithms have been developed the plants. However, it still challenging to build an system with high-quality results at low cost. Useful comparative information existing their improvements also limited, making researchers make data-based selections. objective this study explore possible solutions address...

Recent advances in image-based plant phenotyping have improved our capability to study vegetative stage growth dynamics. However, more complex agronomic traits such as inflorescence architecture (IA), which predominantly contributes grain crop yield are challenging quantify and hence relatively less explored. Previous efforts estimate inflorescence-related using been limited destructive end-point measurements. Development of non-destructive platforms could accelerate the discovery phenotypic...

Abstract Water deficit during the early vegetative growth stages of wheat (Triticum) can limit shoot and ultimately impact grain productivity. Introducing diversity in cultivars to enhance range phenotypic responses water limitations provide potential avenues for mitigating subsequent yield losses. We tested this hypothesis an elite durum background by introducing a series introgressions from wild emmer (Triticum turgidum ssp. dicoccoides) wheat. Wild populations harbor rich drought-adaptive...

Abstract High‐throughput genotyping coupled with molecular breeding approaches have dramatically accelerated crop improvement programs. More recently, improved plant phenotyping methods led to a shift from manual measurements automated platforms increased scalability and resolution. Considerable effort has also gone into developing large‐scale downstream processing of the imaging datasets derived high‐throughput (HTP) platforms. However, most available tools require some programming skills....

Generating 3D digital representations of plants is indispensable for researchers to gain a detailed understanding plant dynamics. Emerging high-throughput phenotyping techniques can capture point clouds that, however, often contain imperfections and make it changeling task generate accurate reconstructions. We present an end-to-end pipeline reconstruct surfaces from maize rice plants. In particular, we propose two-step clustering approach accurately segment the points each individual...

Edge bundling is a promising graph visualization approach to simplifying the visual result of drawing. Plenty edge methods have been developed generate diverse layouts. However, it difficult defend an method with its resulting layout against other as clear theoretic evaluation framework absent in literature. In this paper, we propose information-theoretic evaluate results techniques. We first illustrate advantage visualizations for large graphs, and pinpoint ambiguity from drawing results....

Digital watermarking has been widely used to protect the copyright and integrity of multimedia data. Previous studies mainly focus on designing techniques that are robust attacks destroying embedded watermarks. However, emerging deep learning based image generation technology raises new open issues whether it is possible generate fake watermarked images for circumvention. In this paper, we make first attempt develop digital watermark fakers by using generative adversarial learning. Suppose a...

It is challenging to interpret hyperspectral images in an intuitive and meaningful way, as they usually contain hundreds of dimensions. We develop a visualization tool for based on neural networks, which allows user specify the regions interest, select bands obtain classification results scatterplot generated from features. A cascade network trained generate that matches cluster centers labeled by user. The inferred not only shows clusters points, but also reveals relationships substances....

We have devised and implemented a key technology, SpatioTemporal Adaptive-Resolution Encoding (STARE), in an array database management system, i.e. SciDB, to achieve unparalleled variety scaling for Big Earth Data, enabling rapid-response visual analytics. STARE not only serves as unifying data representation homogenizing diverse varieties of Science Datasets, but also supports spatiotemporal placement alignment these datasets optimize major class analyses, those requiring coincidence. Using...

Studying the growth dynamics of developing plants is critical importance in plant sciences. The traditional methods rely on either manual measurement, which involves tedious labor work, or 2D image-based approaches, cannot fully characterize 3D. Given advances scanners and 3D reconstruction methods, scientists begin to pay more attention models improve accuracy. However, existing mostly focus a whole rather than its detailed substructures. In this paper, we have developed an end-to-end...

With the development of consumer-level depth sensors, 3D face point cloud data can be easily captured now. However, such are often accompanied by low resolution, noise, and holes. At same time, high-precision scanners bulky not widely used in daily applications due to costs inconvenience. To fill gap between high resolution faces, we propose a two-stage framework named super-resolution network (FPSRN) recover high-resolution from low-resolution counterparts. As human faces aligned into...

Deep neural networks conventionally employ end-to-end backpropagation for their training process, which lacks biological credibility and triggers a locking dilemma during network parameter updates, leading to significant GPU memory use. Supervised local learning, segments the into multiple blocks updated by independent auxiliary networks. However, these methods cannot replace due lower accuracy, as gradients only propagate within block, creating lack of information exchange between blocks....