Weather-Domain Transfer-Based Attention YOLO for Multi-Domain Insulator Defect Detection and Classification in UAV Images.

Insulator defect detection of transmission line insulators is an important task for unmanned aerial vehicle (UAV) inspection, which is of immense importance in ensuring the stable operation of transmission lines. Transmission line insulators exist in complex weather scenarios, with small and inconsistent shapes. These insulators under various weather conditions could result in low-quality images captured, limited data numbers, and imbalanced sample problems. Traditional detection methods often struggle to accurately identify defect information, resulting in missed or false detections in real-world scenarios. In this paper, we propose a weather domain synthesis network for extracting cross-modality discriminative information on multi-domain insulator defect detection and classification tasks. Firstly, we design a novel weather domain synthesis (WDSt) module to convert various weather-conditioned insulator images to the uniform weather domain to decrease the existing domain gap. To further improve the detection performance, we leverage the attention mechanism to construct the Cross-modality Information Attention YOLO (CIA-YOLO) model to improve the detection capability for insulator defects. Here, we fuse both shallow and deep feature maps by adding the extra object detection layer, increasing the accuracy for detecting small targets. The experimental results prove the proposed Cross-modality Information Attention YOLO with the weather domain synthesis algorithm can achieve superior performance in multi-domain insulator datasets (MD-Insulator). Moreover, the proposed algorithm also gives a new perspective for decreasing the multi-domain insulator modality gap with weather-domain transfer, which can inspire more researchers to focus on the field.

Identification of potential biomarkers for abdominal pain in IBS patients by bioinformatics approach.

BACKGROUND: Irritable bowel syndrome (IBS) is the most common functional gastrointestinal disease characterized by chronic abdominal discomfort and pain. The mechanisms of abdominal pain, as a relevant symptom, in IBS are still unclear. We aimed to explore the key genes and neurobiological changes specially involved in abdominal pain in IBS. METHODS: Gene expression data (GSE36701) was downloaded from Gene Expression Omnibus database. Fifty-three rectal mucosa samples from 27 irritable bowel syndrome with diarrhea (IBS-D) patients and 40 samples from 21 healthy volunteers as controls were included. Differentially expressed genes (DEGs) between two groups were identified using the GEO2R online tool. Functional enrichment analysis of DEGs was performed on the DAVID database. Then a protein-protein interaction network was constructed and visualized using STRING database and Cytoscape. RESULTS: The microarray analysis demonstrated a subset of genes (CCKBR, CCL13, ACPP, BDKRB2, GRPR, SLC1A2, NPFF, P2RX4, TRPA1, CCKBR, TLX2, MRGPRX3, PAX2, CXCR1) specially involved in pain transmission. Among these genes, we identified GRPR, NPFF and TRPA1 genes as potential biomarkers for irritating abdominal pain of IBS patients. CONCLUSIONS: Overexpression of certain pain-related genes (GRPR, NPFF and TRPA1) may contribute to chronic visceral hypersensitivity, therefore be partly responsible for recurrent abdominal pain or discomfort in IBS patients. Several synapses modification and biological process of psychological distress may be risk factors of IBS.

Bacteria and poisonous plants were the primary causative hazards of foodborne disease outbreak: a seven-year survey from Guangxi, South China.

BACKGROUND: Foodborne diseases are a worldwide public health problem. However, data regarding epidemiological characteristics are still lacking in China. We aimed to analyze the characteristics of foodborne diseases outbreak from 2010 to 2016 in Guangxi, South China. METHODS: A foodborne disease outbreak is the occurrence of two or more cases of a similar foodborne disease resulting from the ingestion of a common food. All data are obtained from reports in the Public Health Emergency Report and Management Information System of the China Information System for Disease Control and Prevention, and also from special investigation reports from Guangxi province. RESULTS: A total of 138 foodborne diseases outbreak occurred in Guangxi in the past 7 years, leading to 3348 cases and 46 deaths. Foodborne disease outbreaks mainly occurred in the second and fourth quarters, and schools and private homes were the most common sites. Ingesting toxic food by mistake, improper cooking and cross contamination were the main routes of poisoning which caused 2169 (64.78%) cases and 37 (80.43%) deaths. Bacteria (62 outbreaks, 44.93%) and poisonous plants (46 outbreaks, 33.33%) were the main etiologies of foodborne diseases in our study. In particular, poisonous plants were the main cause of deaths involved in the foodborne disease outbreaks (26 outbreaks, 56.52%). CONCLUSIONS: Bacteria and poisonous plants were the primary causative hazard of foodborne diseases. Some specific measures are needed for ongoing prevention and control against the occurrence of foodborne diseases.

The fruits of wampee inhibit H2O2-induced apoptosis in PC12 cells via the NF-κB pathway and regulation of cellular redox status.

Wampee (Clausena lansium) fruits (CLS), whose pulp can be used to prepare fruit cups, desserts, jam, or jelly, can be eaten along with the peel. In this study, a PC12 cell model was built to observe the protective effect of CLS against H2O2-induced oxidative stress. We found that pretreatment with CLS increased cell viability and inhibited cytotoxicity, caspase-3 activity and DNA condensation. CLS also attenuated the increase in ROS production and MMP reduction. Moreover, we attempted to determine whether CLS suppressed the expression and phosphorylation of NF-κB. Western blot and immunostaining assay revealed that CLS inhibited H2O2-induced up-regulation of NF-κB p65 and pNF-κB p65. And CLS significantly suppressed the translocation of NF-κB p65 and pNF-κB p65 from cytoplasm to nuclear. Also, seven major compounds including a new flavanoid, luteolin-4'-O-ß-d-gluco-pyranoside (3) and six known compounds 1,2, 4-7 were isolated and identified from CLS. Their antioxidative and H2O2-induced PC12 cell apoptosis-reversing activity were determined. These findings suggest that CLS and its major constituents (flavanoids) may be potential antioxidant agents and should encourage further research into their use as a functional food for neurodegenerative diseases.

Inspector for Face Forgery Detection: Defending Against Adversarial Attacks From Coarse to Fine.

The emergence of face forgery has raised global concerns on social security, thereby facilitating the research on automatic forgery detection. Although current forgery detectors have demonstrated promising performance in determining authenticity, their susceptibility to adversarial perturbations remains insufficiently addressed. Given the nuanced discrepancies between real and fake instances are essential in forgery detection, previous defensive paradigms based on input processing and adversarial training tend to disrupt these discrepancies. For the detectors, the learning difficulty is thus increased, and the natural accuracy is dramatically decreased. To achieve adversarial defense without changing the instances as well as the detectors, a novel defensive paradigm called Inspector is designed specifically for face forgery detectors. Specifically, Inspector defends against adversarial attacks in a coarse-to-fine manner. In the coarse defense stage, adversarial instances with evident perturbations are directly identified and filtered out. Subsequently, in the fine defense stage, the threats from adversarial instances with imperceptible perturbations are further detected and eliminated. Experimental results across different types of face forgery datasets and detectors demonstrate that our method achieves state-of-the-art performances against various types of adversarial perturbations while better preserving natural accuracy. Code is available on https://github.com/xarryon/Inspector.

HiFiSketch: High Fidelity Face Photo-Sketch Synthesis and Manipulation.

With the rapid development of generative adversarial networks, face photo-sketch synthesis has achieved promising performance and playing an increasingly important role in law enforcement as well as entertainment. However, most of the existing methods only work under the condition of no interference, and lack of generalization ability in wild scenes. The fidelity of the images generated by the existing methods are insufficient, and the manipulation ability according to text description is unavailable. Directly applying existing text-based image manipulation methods on face photo-sketch scenario may lead to severe distortions due to the cross-domain challenges. Therefore, we propose a novel cross-domain face photo-sketch synthesis framework named HiFiSketch, a network that learns to adjust the weights of generators for high-fidelity synthesis and manipulation. It can realize the translation of images between the photo domain and the sketch domain, and modify results according to the text input in the meanwhile. We further propose a cross-domain loss function, which can effectively preserve facial details during face photo-sketch synthesis. Extensive experiments on four public face sketch datasets show the superiority of our method compared to existing methods. We further present text-based face photo-sketch manipulation and sequential face photo-sketch manipulation for the first time to demonstrate the effectiveness of our method on high fidelity face photo-sketch synthesis and manipulation.

A review on synthesis and antibacterial potential of bio-selenium nanoparticles in the food industry.

Effective control of foodborne pathogen contamination is a significant challenge to the food industry, but the development of new antibacterial nanotechnologies offers new opportunities. Notably, selenium nanoparticles have been extensively studied and successfully applied in various food fields. Selenium nanoparticles act as food antibacterial agents with a number of benefits, including selenium as an essential trace element in food, prevention of drug resistance induction in foodborne pathogens, and improvement of shelf life and food storage conditions. Compared to physical and chemical methods, biogenic selenium nanoparticles (Bio-SeNPs) are safer and more multifunctional due to the bioactive molecules in Bio-SeNPs. This review includes a summarization of (1) biosynthesized of Bio-SeNPs from different sources (plant extracts, fungi and bacteria) and their antibacterial activity against various foodborne bacteria; (2) the antibacterial mechanisms of Bio-SeNPs, including penetration of cell wall, damage to cell membrane and contents leakage, inhibition of biofilm formation, and induction of oxidative stress; (3) the potential antibacterial applications of Bio-SeNPs as food packaging materials, food additives and fertilizers/feeds for crops and animals in the food industry; and (4) the cytotoxicity and animal toxicity of Bio-SeNPs. The related knowledge contributes to enhancing our understanding of Bio-SeNP applications and makes a valuable contribution to ensuring food safety.

Effect of Laser Remelting Strategy on the Forming Ability of Cemented Carbide Fabricated by Laser Powder Bed Fusion (L-PBF).

Due to the high degree of design freedom and rapid prototyping, laser powder bed fusion (L-PBF) presents a great advantage in the super-hard cemented carbide compared with conventional methods. However, optimizing processing parameters to improve the relative density and surface roughness is still a challenge for cemented carbide fabricated by L-PBF. For this, the effect of the remelting strategy on the forming quality of the L-PBF processed cemented carbide was studied in this article, aiming to explore a suitable process window. The surface quality, relative density, microstructure, and microhardness of the cemented carbide parts fabricated under a single melting and remelting strategy were compared. The results showed that the remelting strategy could efficiently improve the specimens' surface quality and relative density. Besides, the cracks were not obviously aggravated, and the WC grains could distribute more homogeneously on the binder matrix under the remelting strategy. Therefore, the microhardness showed an improvement compared to the single melting strategy.

Heterogeneous Face Interpretable Disentangled Representation for Joint Face Recognition and Synthesis.

Heterogeneous faces are acquired with different sensors, which are closer to real-world scenarios and play an important role in the biometric security field. However, heterogeneous face analysis is still a challenging problem due to the large discrepancy between different modalities. Recent works either focus on designing a novel loss function or network architecture to directly extract modality-invariant features or synthesizing the same modality faces initially to decrease the modality gap. Yet, the former always lacks explicit interpretability, and the latter strategy inherently brings in synthesis bias. In this article, we explore to learn the plain interpretable representation for complex heterogeneous faces and simultaneously perform face recognition and synthesis tasks. We propose the heterogeneous face interpretable disentangled representation (HFIDR) that could explicitly interpret dimensions of face representation rather than simple mapping. Benefited from the interpretable structure, we further could extract latent identity information for cross-modality recognition and convert the modality factor to synthesize cross-modality faces. Moreover, we propose a multimodality heterogeneous face interpretable disentangled representation (M-HFIDR) to extend the basic approach suitable for the multimodality face recognition and synthesis. To evaluate the ability of generalization, we construct a novel large-scale face sketch data set. Experimental results on multiple heterogeneous face databases demonstrate the effectiveness of the proposed method.

Effects of Different Substrates on the Formability and Densification Behaviors of Cemented Carbide Processed by Laser Powder Bed Fusion.

Cemented carbide materials are widely applied in cutting tools, drill tools, and mold fabrication due to their superior hardness and wear resistance. Producing cemented carbide parts via the laser powder bed fusion (L-PBF) method has the advantage of fabricating complex structures with a rapid manufacturing speed; however, they were underdeveloped due to their low density and crack formation on the blocks. This work studied the effect of different substrates including 316L substrates, Ni200 substrates, and YG15 substrates on the forming quality of WC-17Co parts fabricated by L-PBF, with the aim of finding the optimal substrate for fabrication. The results revealed that the Ni200 substrates had a better wettability for the single tracks formation than other substrates, and bonding between the built block and the Ni200 substrate was firm without separation during processing with a large range of laser energy inputs. This guaranteed the fabrication of a relatively dense block with fewer cracks. Although the high laser energy input that led to fine crack formation on the blocks formed on the Ni200 substrate, it was found to be better suited to restricting cracks than other substrates.

Moderately Crystalline Azine-Linked Covalent Organic Framework Membrane for Ultrafast Molecular Sieving.

Covalent organic frameworks are potential candidates for the preparation of advanced molecular separation membranes due to their porous structure, uniform aperture, and chemical stability. However, the fabrication of continuous COF membranes in a facile and mild manner remains a challenge. Herein, a continuous, defect-free, and flexible azine-linked ACOF-1 membrane was prepared on a hydrolyzed polyacrylonitrile (HPAN) substrate via in situ interfacial polymerization (IP). A moderately crystalline COF ultrathin selective layer enabled ultrafast molecular sieving. The effect of synthesis parameters including precursor concentration, catalyst dosage, and reaction duration on the dye separation performance was investigated. The optimized membrane displayed an ultrahigh water permeance of 142 L m-2 h-1 bar-1 together with favorable rejection (e.g., 99.2% for Congo red and 96.3% for methyl blue). The water permeance is 5-12 times higher than that of reported membranes with similar rejections. In addition, ACOF-1 membranes demonstrate outstanding long-term stability together with organic solvent and extreme pH resistance. Meanwhile, the membrane is suitable for removing dyes from salt solution products owing to their nonselective permeation for hydrated salt ions (<10.6%). The superior performance and the excellent chemical stability render the ACOF-1 membrane a satisfactory system for water purification.

Coupled Attribute Learning for Heterogeneous Face Recognition.

Heterogeneous face recognition (HFR) is a challenging problem in face recognition and subject to large textural and spatial structure differences of face images. Different from conventional face recognition in homogeneous environments, there exist many face images taken from different sources (including different sensors or different mechanisms) in reality. In addition, limited training samples of cross-modality pairs make HFR more challenging due to the complex generation procedure of these images. Despite the great progress that has been achieved in recent years, existing works mainly focus on HFR from only cross-modality image matching. However, it is more practical to obtain both facial images and semantic descriptions about facial attributes in real-world situations, in which the semantic description clues are nearly always obtained during the process of image generation. Motivated by human cognitive mechanisms, we naturally utilize the explicit invariant semantic description, i.e., face attributes, to help address the gap among face images of different modalities. Existing facial attributes-related face recognition methods primarily regard attributes as the high-level features used to enhance recognition performance, ignoring the inherent relationship between face attributes and identities. In this article, we propose novel coupled attribute learning for the HFR (CAL-HFR) method without labeling the attributes manually. Deep convolutional networks are employed to directly map face images in heterogeneous scenarios to a compact common space where distances are taken as dissimilarities of pairs. Coupled attribute guided triplet loss (CAGTL) is designed to train an end-to-end HFR network that can effectively eliminate defects of incorrectly estimated attributes. Extensive experiments on multiple heterogeneous scenarios demonstrate that the proposed method achieves superior performance compared with that of state-of-the-art methods. Furthermore, we make publicly available our generated pairwise annotated heterogeneous facial attribute database for evaluation and promoting related research.