Application of amodal segmentation for shape reconstruction and occlusion recovery in occluded tomatoes.

Common object detection and image segmentation methods are unable to accurately estimate the shape of the occluded fruit. Monitoring the growth status of shaded crops in a specific environment is challenging, and certain studies related to crop harvesting and pest detection are constrained by the natural shadow conditions. Amodal segmentation can focus on the occluded part of the fruit and complete the overall shape of the fruit. We proposed a Transformer-based amodal segmentation algorithm to infer the amodal shape of occluded tomatoes. Considering the high cost of amodal annotation, we only needed modal dataset to train the model. The dataset was taken from two greenhouses on the farm and contains rich occlusion information. We introduced boundary estimation in the hourglass structured network to provide a priori information about the completion of the amodal shapes, and reconstructed the occluded objects using a GAN network (with discriminator) and GAN loss. The model in this study showed accuracy, with average pairwise accuracy of 96.07%, mean intersection-over-union (mIoU) of 94.13% and invisible mIoU of 57.79%. We also examined the quality of pseudo-amodal annotations generated by our proposed model using Mask R-CNN. Its average precision (AP) and average precision with intersection over union (IoU) 0.5 (AP50) reached 63.91%,86.91% respectively. This method accurately and rationally achieves the shape of occluded tomatoes, saving the cost of manual annotation, and is able to deal with the boundary information of occlusion while decoupling the relationship of occluded objects from each other. Future work considers how to complete the amodal segmentation task without overly relying on the occlusion order and the quality of the modal mask, thus promising applications to provide technical support for the advancement of ecological monitoring techniques and ecological cultivation.

Directional and Orderly Arranged Ni0.9Mn0.1(OH)2 Enables the Synthesis of Single-Crystal Ni-Rich Co-Free LiNi0.9Mn0.1O2 with Enhanced Internal Structural Stability.

In this paper, the effect of the structure characteristics of the precursor on the electrochemical properties of a single-crystal cobalt-free high-nickel LiNi0.9Mn0.1O2 cathode is systematically studied. Precursors with different morphologies are synthesized by adjusting the coprecipitation reaction conditions. The results of SEM and XRD show that with the increase in the orderly stacking arrangement of internal primary nanosheets of Ni0.9Mn0.1(OH)2, the exposed active {010} planes at the surface increase. The prepared cathode materials finally inherit the structural features of the precursor, and the single-crystal Co-free Ni-rich LiNi0.9Mn0.1O2 cathode with highly exposed active {010} planes shows a well-ordered crystal structure and low Li+/Ni2+ cation mixing. The characterization results reveal that the high percentage of {010} planes will improve the Li+ transportation kinetics, decrease electrochemical impedance, and significantly alleviate the accumulation of rock-salt phases. Therefore, the material with this structure shows good electrochemical performance.

Cellulose-Assisted Vertically Heterostructured PEO-Based Solid Electrolytes Mitigating Li-Succinonitrile Corrosion for Lithium Metal Batteries.

In the field of solid-state lithium metal batteries (SSLMBs), constructing vertically heterostructured poly(ethylene oxide) (PEO)-based solid electrolytes is an effective method to realize their tight contact with cathodes and Li anodes at the same time. Succinonitrile (SN) has been widely used in PEO-based solid electrolytes to improve the interface contact with cathodes, enhance the ionic conductivities, and obtain a high electrochemical stability window of PEO, but its application is still hindered by its intrinsic instability to Li anodes, which results in corrosion and side interactions with lithium metal. Herein, the cellulose membrane (CM) is introduced creatively into the vertically heterostructured PEO-based solid electrolytes to match the PEO-SN solid electrolytes at the cathode side. With the advantage of the interaction between -OH groups of CM and -C≡N groups in SN, the movement of free SN molecules from cathodes to Li anodes is limited effectively, resulting in a stable and durable SEI layer. In specific, the Li||LiFePO4 battery with the CM-assisted vertically heterostructured PEO-based solid electrolyte by in situ preparation delivers a discharge capacity of around 130 mAh g-1 after 300 cycles and capacity retention of 95% after 500 cycles at 0.5 C. Our work provides a solution to construct PEO-based solid electrolytes feasible to match cathodes and Li anodes effectively by intimate contact with electrodes.

DFSP: A fast and automatic distance field-based stem-leaf segmentation pipeline for point cloud of maize shoot.

The 3D point cloud data are used to analyze plant morphological structure. Organ segmentation of a single plant can be directly used to determine the accuracy and reliability of organ-level phenotypic estimation in a point-cloud study. However, it is difficult to achieve a high-precision, automatic, and fast plant point cloud segmentation. Besides, a few methods can easily integrate the global structural features and local morphological features of point clouds relatively at a reduced cost. In this paper, a distance field-based segmentation pipeline (DFSP) which could code the global spatial structure and local connection of a plant was developed to realize rapid organ location and segmentation. The terminal point clouds of different plant organs were first extracted via DFSP during the stem-leaf segmentation, followed by the identification of the low-end point cloud of maize stem based on the local geometric features. The regional growth was then combined to obtain a stem point cloud. Finally, the instance segmentation of the leaf point cloud was realized using DFSP. The segmentation method was tested on 420 maize and compared with the manually obtained ground truth. Notably, DFSP had an average processing time of 1.52 s for about 15,000 points of maize plant data. The mean precision, recall, and micro F1 score of the DFSP segmentation algorithm were 0.905, 0.899, and 0.902, respectively. These findings suggest that DFSP can accurately, rapidly, and automatically achieve maize stem-leaf segmentation tasks and could be effective in maize phenotype research. The source code can be found at https://github.com/syau-miao/DFSP.git.

Li-Deficient Materials-Decoration Restrains Oxygen Evolution Achieving Excellent Cycling Stability of Li-Rich Mn-Based Cathode.

With the increasing demand for high energy density and rapid charging performance, Li-rich materials have been the up and coming cathodes for next-generation lithium-ion batteries. However, because of oxygen evolution and structural instability, the commercialization of Li-rich materials is extremely retarded by their poor electrochemical performances. In this work, Li-deficient materials Li0.3NbO2 and (Nb0.62Li0.15)TiO3 are applied to functionalize the surface of Li1.2Mn0.54Ni0.13Co0.13O2, aiming to suppress oxygen evolution and increase structural stability in LIBs. In addition, a fast Li-ion transport channel is beneficial to enhance Li+ diffusion kinetics. The results demonstrate that the electrodes decorated with Li0.3NbO2 and (Nb0.62Li0.15)TiO3 materials exhibit more stable cycling stability after long-term cycling and outstanding rate capability.

Deep learning methods for automatic evaluation of delayed enhancement-MRI. The results of the EMIDEC challenge.

A key factor for assessing the state of the heart after myocardial infarction (MI) is to measure whether the myocardium segment is viable after reperfusion or revascularization therapy. Delayed enhancement-MRI or DE-MRI, which is performed 10 min after injection of the contrast agent, provides high contrast between viable and nonviable myocardium and is therefore a method of choice to evaluate the extent of MI. To automatically assess myocardial status, the results of the EMIDEC challenge that focused on this task are presented in this paper. The challenge's main objectives were twofold. First, to evaluate if deep learning methods can distinguish between non-infarct and pathological exams, i.e. exams with or without hyperenhanced area. Second, to automatically calculate the extent of myocardial infarction. The publicly available database consists of 150 exams divided into 50 cases without any hyperenhanced area after injection of a contrast agent and 100 cases with myocardial infarction (and then with a hyperenhanced area on DE-MRI), whatever their inclusion in the cardiac emergency department. Along with MRI, clinical characteristics are also provided. The obtained results issued from several works show that the automatic classification of an exam is a reachable task (the best method providing an accuracy of 0.92), and the automatic segmentation of the myocardium is possible. However, the segmentation of the diseased area needs to be improved, mainly due to the small size of these areas and the lack of contrast with the surrounding structures.

Surface LiMn1.4Ni0.5Mo0.1O4 Coating and Bulk Mo Doping of Li-Rich Mn-Based Li1.2Mn0.54Ni0.13Co0.13O2 Cathode with Enhanced Electrochemical Performance for Lithium-Ion Batteries.

To improve the initial Coulombic efficiency, cycling stability, and rate performance of the Li-rich Mn-based Li1.2Mn0.54Ni0.13Co0.13O2 cathode, the combination of LiMn1.4Ni0.5Mo0.1O4 coating with Mo doping has been successfully carried out by the sol-gel method and subsequent dip-dry process. This strategy buffers the electrodes from the corrosion of electrolyte and enhances the lattice parameter, which could inhibit the oxygen release and maintain the structural stability, thus improving the cycle stability and rate capability. After LiMn1.4Ni0.5Mo0.1O4 modification, the initial discharge capacity reaches 272.4 mAh g-1 with a corresponding initial Coulombic efficiency (ICE) of 84.2% at 0.1C (1C = 250 mAh g-1), far higher than those (221.5 mAh g-1 and 68.9%) of the pristine sample. Besides, the capacity retention of the coated sample is enhanced by up to 66.8% after 200 cycles at 0.1C. Especially, the rate capability of the coated sample is 95.2 mAh g-1 at 5C. XRD, SEM, TEM, XPS, and Raman spectroscopy are adopted to characterize the morphologies and structures of the samples. This coating strategy has been demonstrated to be an effective approach to construct high-performance energy storage devices.

Depth Density Achieves a Better Result for Semantic Segmentation with the Kinect System.

Image segmentation is one of the most important methods for animal phenome research. Since the advent of deep learning, many researchers have looked at multilayer convolutional neural networks to solve the problems of image segmentation. A network simplifies the task of image segmentation with automatic feature extraction. Many networks struggle to output accurate details when dealing with pixel-level segmentation. In this paper, we propose a new concept: Depth density. Based on a depth image, produced by a Kinect system, we design a new function to calculate the depth density value of each pixel and bring this value back to the result of semantic segmentation for improving the accuracy. In the experiment, we choose Simmental cattle as the target of image segmentation and fully convolutional networks (FCN) as the verification networks. We proved that depth density can improve four metrics of semantic segmentation (pixel accuracy, mean accuracy, mean intersection over union, and frequency weight intersection over union) by 2.9%, 0.3%, 11.4%, and 5.02%, respectively. The result shows that depth information produced by Kinect can improve the accuracy of the semantic segmentation of FCN. This provides a new way of analyzing the phenotype information of animals.

A comparative study on membrane fouling alleviation mechanisms by using nanoscale Fe3O4 and poly dimethyldiallylammonium chloride.

Membrane bioreactor (MBR) has become a promising technology for wastewater treatment. However, membrane fouling frequently occurred which greatly increased operational expense. Two different membrane fouling alleviation mechanisms were explored in this study. Addition of poly dimethyldiallylammonium chloride (PDMDAAC) facilitated formation of flocs-flocs aggregates, which were more adaptable to the changing environment, resulting in less soluble microbial products (SMP) secretion. However, PDMDAAC lose activity gradually, and had a less sustainable effect on membrane fouling alleviation. Nanoscale Fe3O4 was applied to alleviate membrane fouling, and membrane sustainable filtration cycle extended 2-fold compared to the control group. Results showed that dehydrogenase activity in the reactor with optimal addition of nanoscale Fe3O4 increased 2.86 ± 0.11 times compared to control group. SMP (especially tryptophan protein-like substances) decreased to 9.79 ± 1.34 mg L-1 with the addition of nanoscale Fe3O4, which was lower than that in the control group (15.31 ± 0.53 mg L-1). It's speculated that nanoscale Fe3O4 performed as conductive material, which intensified interspecies electron transfer. The sludge dehydrogenase activity was then enhanced, which facilitated the utilization and microbial degradation of SMP, suppressing membrane fouling consequently.

Diagnosis of Alzheimer's Disease via Multi-Modality 3D Convolutional Neural Network.

Alzheimer's disease (AD) is one of the most common neurodegenerative diseases. In the last decade, studies on AD diagnosis has attached great significance to artificial intelligence-based diagnostic algorithms. Among the diverse modalities of imaging data, T1-weighted MR and FDG-PET are widely used for this task. In this paper, we propose a convolutional neural network (CNN) to integrate all the multi-modality information included in both T1-MR and FDG-PET images of the hippocampal area, for the diagnosis of AD. Different from the traditional machine learning algorithms, this method does not require manually extracted features, instead, it utilizes 3D image-processing CNNs to learn features for the diagnosis or prognosis of AD. To test the performance of the proposed network, we trained the classifier with paired T1-MR and FDG-PET images in the ADNI datasets, including 731 cognitively unimpaired (labeled as CN) subjects, 647 subjects with AD, 441 subjects with stable mild cognitive impairment (sMCI) and 326 subjects with progressive mild cognitive impairment (pMCI). We obtained higher accuracies of 90.10% for CN vs. AD task, 87.46% for CN vs. pMCI task, and 76.90% for sMCI vs. pMCI task. The proposed framework yields a state-of-the-art performance. Finally, the results have demonstrated that (1) segmentation is not a prerequisite when using a CNN for the classification, (2) the combination of two modality imaging data generates better results.

Population Genomic Analysis and De Novo Assembly Reveal the Origin of Weedy Rice as an Evolutionary Game.

Crop weediness, especially that of weedy rice (Oryza sativa f. spontanea), remains mysterious. Weedy rice possesses robust ecological adaptability; however, how this strain originated and gradually formed proprietary genetic features remains unclear. Here, we demonstrate that weedy rice at Asian high latitudes (WRAH) is phylogenetically well defined and possesses unselected genomic characteristics in many divergence regions between weedy and cultivated rice. We also identified novel quantitative trait loci underlying weedy-specific traits, and revealed that a genome block on the end of chromosome 1 is associated with rice weediness. To identify the genomic modifications underlying weedy rice evolution, we generated the first de novo assembly of a high-quality weedy rice genome (WR04-6), and conducted a comparative genomics study between WR04-6 with other rice reference genomes. Multiple lines of evidence, including the results of demographic scenario comparisons, suggest that differentiation between weedy rice and cultivated rice was initiated by genetic improvement of cultivated rice and that the essence of weediness arose through semi-domestication. A plant height model further implied that the origin of WRAH can be modeled as an evolutionary game and indicated that strategy-based selection driven by fitness shaped its genomic diversity.

Degradation of pentachlorobiphenyl by a sequential treatment using Pd coated iron and an aerobic bacterium (H1).

The reductive dechlorination and biodegradation of 2,2(')4,5,5(')-pentachlorobiphenyl (PCB#101) was investigated in a laboratory-scale. Palladium coated iron (Pd/Fe) was used as a catalytic reductant for the chemical degradation of 2,2(')4,5,5(')-pentachlorobiphenyl, and an aerobic bacteria was used for biodegradation following the chemical reaction in this study. Dechlorination was affected by several factors such as Pd loading, initial soil pH and the amount of Pd/Fe used. The results showed that higher Pd loading, higher dosage of Pd/Fe and slightly acid condition were beneficial to the catalytic dechlorination of 2,2('),4,5,5(')-pentachlorobiphenyl. In laboratory batch experiments, 2,2(')4,5,5(')-pentachlorobiphenyl was reduced in the presence of Pd/Fe bimetal, which was not further degraded by aerobic bacteria. 2,2('),4-trichlorobiphenyl (PCB#17), a reduction product from 2,2(')4,5,5(')-pentachlorobiphenyl, was readily biodegraded in the presence of a aerobic bacterial strain. It is suggested that an integrated Pd/Fe catalytic reduction-aerobic biodegradation process may be a feasible option for treating PCB-contaminated soil.

Catalytic dechlorination of polychlorinated biphenyls in soil by palladium-iron bimetallic catalyst.

Pd/Fe bimetallic particles were synthesized by chemical deposition and used to dechlorinate 2,2',4,5,5'-pentachlorobiphenyl in soil. Batch experiments demonstrated that the Pd/Fe bimetallic particles could effectively dechlorinate 2,2',4,5,5'-pentachlorobiphenyl. Dechlorination was affected by several factors such as reaction time, Pd loading, the amount of Pd/Fe used, initial soil pH, and 2,2',4,5,5'-pentachlorobiphenyl concentration. The results showed that higher Pd loading, higher dosage of Pd/Fe, lower initial concentration of 2,2',4,5,5'-pentachlorobiphenyl and slightly acid condition were beneficial to the catalytic dechlorination of 2,2',4,5,5'-pentachlorobiphenyl. The degradation of 2,2',4,5,5'-pentachlorobiphenyl, catalyzed by Pd/Fe followed pseudo-first-order kinetics.