Object and spatial discrimination makes weakly supervised local feature better.

Local feature extraction plays a crucial role in numerous critical visual tasks. However, there remains room for improvement in both descriptors and keypoints, particularly regarding the discriminative power of descriptors and the localization precision of keypoints. To address these challenges, this study introduces a novel local feature extraction pipeline named OSDFeat (Object and Spatial Discrimination Feature). OSDFeat employs a decoupling strategy, training descriptor and detection networks independently. Inspired by semantic correspondence, we propose an Object and Spatial Discrimination ResUNet (OSD-ResUNet). OSD-ResUNet captures features from the feature map that differentiate object appearance and spatial context, thus enhancing descriptor performance. To further improve the discriminative capability of descriptors, we propose a Discrimination Information Retained Normalization module (DIRN). DIRN complementarily integrates spatial-wise normalization and channel-wise normalization, yielding descriptors that are more distinguishable and informative. In the detection network, we propose a Cross Saliency Pooling module (CSP). CSP employs a cross-shaped kernel to aggregate long-range context in both vertical and horizontal dimensions. By enhancing the saliency of keypoints, CSP enables the detection network to effectively utilize descriptor information and achieve more precise localization of keypoints. Compared to the previous best local feature extraction methods, OSDFeat achieves Mean Matching Accuracy of 79.4% in local feature matching task, improving by 1.9% and achieving state-of-the-art results. Additionally, OSDFeat achieves competitive results in Visual Localization and 3D Reconstruction. The results of this study indicate that object and spatial discrimination can improve the accuracy and robustness of local feature, even in challenging environments. The code is available at https://github.com/pandaandyy/OSDFeat.

Illumination-aware divide-and-conquer network for improperly-exposed image enhancement.

Improperly-exposed images often have unsatisfactory visual characteristics like inadequate illumination, low contrast, and the loss of small structures and details. The mapping relationship from an improperly-exposed condition to a well-exposed one may vary significantly due to the presence of multiple exposure conditions. Consequently, the enhancement methods that do not pay specific attention to this issue tend to yield inconsistent results when applied to the same scene under different exposure conditions. In order to obtain consistent enhancement results for various exposures while restoring rich details, we propose an illumination-aware divide-and-conquer network (IDNet). Specifically, to address the challenge of directly learning a sophisticated nonlinear mapping from an improperly-exposed condition to a well-exposed one, we utilize the discrete wavelet transform (DWT) to decompose the image into the low-frequency (LF) component, which primarily captures brightness and contrast, and the high-frequency (HF) components that depict fine-scale structures. To mitigate the inconsistency in correction across various exposures, we extract a conditional feature from the input that represents illumination-related global information. This feature is then utilized to modulate the dynamic convolution weights, enabling precise correction of the LF component. Furthermore, as the co-located positions of LF and HF components are highly correlated, we create a mask to distill useful knowledge from the corrected LF component, and integrate it into the HF component to support the restoration of fine-scale details. Extensive experimental results demonstrate that the proposed IDNet is superior to several state-of-the-art enhancement methods on two datasets with multiple exposures.

Diversity and Pathogenicity of Fusarium Root Rot Fungi from Canola (Brassica napus) in Alberta, Canada.

Root rot disease poses a significant threat to canola (Brassica napus), underscoring the need for a comprehensive understanding of its causal agents for more effective disease mitigation. The composition and diversity of fungal pathogens associated with root rot of canola in Alberta, Canada, were evaluated from plant tissue samples collected in 2021 and 2022. The study revealed Fusarium spp. as the predominant pathogens found in almost all surveyed fields. Fusarium avenaceum, F. redolens, and F. solani were among the most frequently recovered species. Greenhouse trials confirmed their pathogenicity, with F. avenaceum and F. sporotrichioides found to be particularly aggressive. Additionally, F. sporotrichioides and F. commune were identified for the first time as canola root rot pathogens. Inoculation with isolates of most species resulted in significant reductions in seedling emergence, plant height, and shoot and root dry weights. Analysis of translation elongation factor 1-α (TEF-1α) and internal transcribed spacer (ITS) sequences confirmed the identity of the Fusarium spp., while concatenating the ITS and TEF-1α sequences enabled improved species differentiation. Geographic and year effects did not influence fungal diversity or aggressiveness, as determined by principal component analysis. This study emphasized the high diversity and impact of Fusarium spp. in causing canola root rot.

FD-Net: Feature Distillation Network for Oral Squamous Cell Carcinoma Lymph Node Segmentation in Hyperspectral Imagery.

Oral squamous cell carcinoma (OSCC) has the characteristics of early regional lymph node metastasis. OSCC patients often have poor prognoses and low survival rates due to cervical lymph metastases. Therefore, it is necessary to rely on a reasonable screening method to quickly judge the cervical lymph metastastic condition of OSCC patients and develop appropriate treatment plans. In this study, the widely used pathological sections with hematoxylin-eosin (H&E) staining are taken as the target, and combined with the advantages of hyperspectral imaging technology, a novel diagnostic method for identifying OSCC lymph node metastases is proposed. The method consists of a learning stage and a decision-making stage, focusing on cancer and non-cancer nuclei, gradually completing the lesions' segmentation from coarse to fine, and achieving high accuracy. In the learning stage, the proposed feature distillation-Net (FD-Net) network is developed to segment the cancerous and non-cancerous nuclei. In the decision-making stage, the segmentation results are post-processed, and the lesions are effectively distinguished based on the prior. Experimental results demonstrate that the proposed FD-Net is very competitive in the OSCC hyperspectral medical image segmentation task. The proposed FD-Net method performs best on the seven segmentation evaluation indicators: MIoU, OA, AA, SE, CSI, GDR, and DICE. Among these seven evaluation indicators, the proposed FD-Net method is 1.75%, 1.27%, 0.35%, 1.9%, 0.88%, 4.45%, and 1.98% higher than the DeepLab V3 method, which ranks second in performance, respectively. In addition, the proposed diagnosis method of OSCC lymph node metastasis can effectively assist pathologists in disease screening and reduce the workload of pathologists.

Multi-scale feature selection network for lightweight image super-resolution.

Recently, many super-resolution (SR) methods based on convolutional neural networks (CNNs) have achieved superior performance by utilizing deep and heavy models, which may not be suitable for real-world low-budget devices. To address this issue, we propose a novel lightweight SR network called a multi-scale feature selection network (MFSN). As the basic building block of MFSN, the multi-scale feature selection block (MFSB) is presented to extract the rich multi-scale features from a coarse-to-fine receptive field level. For a better representation ability, a wide-activated residual unit is adopted in each branch of MFSB except the last one. In MFSB, the scale selection module (SSM) is designed to effectively fuse the features from two adjacent branches by adjusting receptive field sizes adaptively. Further, a comprehensive channel attention mechanism (CCAM) is integrated into SSM to learn the dynamic selection weight by considering the local and global inter-channel dependencies. Extensive experimental results illustrate that the proposed MFSN is superior to other lightweight methods.

Characterization of the Virulence and Yield Impact of Fusarium Species on Canola (Brassica napus).

Multiple species of Fusarium can contribute to the development of root rot in canola (Brassica napus), making disease management difficult. We conducted field and greenhouse experiments to investigate the impacts of Fusarium avenaceum and Fusarium proliferatum, and the interaction between Fusarium oxysporum and F. proliferatum on root rot severity and canola yields. Inoculation with any of the three Fusarium spp. resulted in significant disease severity and reduced seedling emergence compared with non-inoculated controls, leading to yield reductions of up to 35%. Notably, there was a strong correlation (r = 0.93) between root rot severity at the seedling stage and at maturity. Regression analysis indicated a linear decline in seedling emergence with increasing disease severity. Furthermore, disease severity at maturity adversely affected the pod number per plant and the seed weight per plant, with both parameters ultimately approaching zero at a severity of 4.0 on a 0-4 scale. Co-inoculation with F. oxysporum and F. proliferatum induced more severe root rot than inoculation with each species on its own, suggesting synergistic interactions between these fungi. Knowledge of these interactions and the relative virulence of Fusarium spp. will contribute to the improved management of root rot in canola.

DSCA-PSPNet: Dynamic spatial-channel attention pyramid scene parsing network for sugarcane field segmentation in satellite imagery.

Sugarcane plays a vital role in many global economies, and its efficient cultivation is critical for sustainable development. A central challenge in sugarcane yield prediction and cultivation management is the precise segmentation of sugarcane fields from satellite imagery. This task is complicated by numerous factors, including varying environmental conditions, scale variability, and spectral similarities between crops and non-crop elements. To address these segmentation challenges, we introduce DSCA-PSPNet, a novel deep learning model with a unique architecture that combines a modified ResNet34 backbone, the Pyramid Scene Parsing Network (PSPNet), and newly proposed Dynamic Squeeze-and-Excitation Context (D-scSE) blocks. Our model effectively adapts to discern the importance of both spatial and channel-wise information, providing superior feature representation for sugarcane fields. We have also created a comprehensive high-resolution satellite imagery dataset from Guangxi's Fusui County, captured on December 17, 2017, which encompasses a broad spectrum of sugarcane field characteristics and environmental conditions. In comparative studies, DSCA-PSPNet outperforms other state-of-the-art models, achieving an Intersection over Union (IoU) of 87.58%, an accuracy of 92.34%, a precision of 93.80%, a recall of 93.21%, and an F1-Score of 92.38%. Application tests on an RTX 3090 GPU, with input image resolutions of 512 × 512, yielded a prediction time of 4.57ms, a parameter size of 22.57MB, GFLOPs of 11.41, and a memory size of 84.47MB. An ablation study emphasized the vital role of the D-scSE module in enhancing DSCA-PSPNet's performance. Our contributions in dataset generation and model development open new avenues for tackling the complexities of sugarcane field segmentation, thus contributing to advances in precision agriculture. The source code and dataset will be available on the GitHub repository https://github.com/JulioYuan/DSCA-PSPNet/tree/main.

Identification of Novel Genes Associated with Partial Resistance to Aphanomyces Root Rot in Field Pea by BSR-Seq Analysis.

Aphanomyces root rot, caused by Aphanomyces euteiches, causes severe yield loss in field pea (Pisum sativum). The identification of a pea germplasm resistant to this disease is an important breeding objective. Polygenetic resistance has been reported in the field pea cultivar '00-2067'. To facilitate marker-assisted selection (MAS), bulked segregant RNA-seq (BSR-seq) analysis was conducted using an F8 RIL population derived from the cross of 'Carman' × '00-2067'. Root rot development was assessed under controlled conditions in replicated experiments. Resistant (R) and susceptible (S) bulks were constructed based on the root rot severity in a greenhouse study. The BSR-seq analysis of the R bulks generated 44,595,510~51,658,688 reads, of which the aligned sequences were linked to 44,757 genes in a reference genome. In total, 2356 differentially expressed genes were identified, of which 44 were used for gene annotation, including defense-related pathways (jasmonate, ethylene and salicylate) and the GO biological process. A total of 344.1 K SNPs were identified between the R and S bulks, of which 395 variants were located in 31 candidate genes. The identification of novel genes associated with partial resistance to Aphanomyces root rot in field pea by BSR-seq may facilitate efforts to improve management of this important disease.

Mapping QTL associated with partial resistance to Aphanomyces root rot in pea (Pisum sativum L.) using a 13.2 K SNP array and SSR markers.

KEY MESSAGE: A stable and major QTL, which mapped to an approximately 20.0 cM region on pea chromosome 4, was identified as the most consistent region conferring partial resistance to Aphanomyces euteiches. Aphanomyces root rot (ARR), caused by Aphanomyces euteiches Drechs., is a destructive soilborne disease of field pea (Pisum Sativum L.). No completely resistant pea germplasm is available, and current ARR management strategies rely on partial resistance and fungicidal seed treatments. In this study, an F8 recombinant inbred line population of 135 individuals from the cross 'Reward' (susceptible) × '00-2067' (tolerant) was evaluated for reaction to ARR under greenhouse conditions with the A. euteiches isolate Ae-MDCR1 and over 2 years in a field nursery in Morden, Manitoba. Root rot severity, foliar weight, plant vigor and height were used as estimates of tolerance to ARR. Genotyping was conducted with a 13.2 K single-nucleotide polymorphism (SNP) array and 222 simple sequence repeat (SSR) markers. Statistical analyses of the phenotypic data indicated significant (P < 0.001) genotypic effects and significant G × E interactions (P < 0.05) in all experiments. After filtering, 3050 (23.1%) of the SNP and 30 (13.5%) of the SSR markers were retained for linkage analysis, which distributed 2999 (2978 SNP + 21 SSR) of the markers onto nine linkage groups representing the seven chromosomes of pea. Mapping of quantitative trait loci (QTL) identified 8 major-effect (R2 > 20%), 13 moderate-effect (10% < R2 < 20%) effect and 6 minor-effect (R2 < 10%) QTL. A genomic region on chromosome 4, delimited by the SNP markers PsCam037549_22628_1642 and PsCam026054_14999_2864, was identified as the most consistent region responsible for partial resistance to A. euteiches isolate Ae-MDCR1. Other genomic regions important for resistance were of the order chromosome 5, 6 and 7.

Identification of Quantitative Trait Loci Associated With Partial Resistance to Fusarium Root Rot and Wilt Caused by Fusarium graminearum in Field Pea.

Fusarium root rot, caused by a complex of Fusarium spp., is a major disease of field pea (Pisum sativum). The development of genetic resistance is the most promising approach to manage the disease, but no pea germplasm has been identified that is completely resistant to root rot. The aim of this study was to detect quantitative trait loci (QTL) conferring partial resistance to root rot and wilting, caused by five fungal isolates representing Fusarium solani, F. avenaceum, F. acuminatum, F. proliferatum, and F. graminearum. Evaluation of the root rot-tolerant cultivar "00-2067" and susceptible cultivar "Reward" was carried out with the five species. There was a significant difference (p < 0.001) between the mean root rot values of the two cultivars inoculated with the F. avenaceum (F4A) and F. graminearum (FG2) isolates. Therefore, in the QTL study, the F8 recombinant inbred line (RIL) population derived from "Reward" × "00-2067" was inoculated in the greenhouse (4 ×) with only F4A and FG2. The parents and F8 population were genotyped using 13.2K single nucleotide polymorphisms (SNPs) and 222 simple sequence repeat (SSR) markers. A significant genotypic effect (p < 0.05) and high heritability (79% to 92.1%) were observed for disease severity, vigor, and plant height following inoculation with F4A and FG2. Significant correlation coefficients were detected among and within all traits. This suggested that a high proportion of the genetic variance was transmitted from the parents to the progeny. However, no significant QTL (LOD > 3) were detected for the RILs inoculated with F4A. In the case of the RILs inoculated with FG2, 5 QTL for root rot severity and 3 QTL each for vigor and plant height were detected. The most stable QTL for plant height (Hgt-Ps3.1) was detected on Chrom5/LGIII. The two most stable QTL for partial resistance to FG2, Fg-Ps4.1, and Fg-Ps4.2 were located in a 15.1-cM and 11.2-cM genomic region, respectively, on Chrom4/LGIV. The most stable QTL for vigor (Vig-Ps4.1) was found in the same region. Twenty-five major and moderate effect digenic epistatic interactions were detected. The identified region on chrom4/LGIV could be important for resistance breeding and marker development.

Neonatal Dexamethasone Treatment Exacerbates Hypoxia/Ischemia-Induced White Matter Injury.

Dexamethasone, a synthetic glucocorticoid, has been widely used to prevent or ameliorate morbidity of chronic lung disease in preterm infants with respiratory distress syndrome. Despite its beneficial effect on neonatal lung function, growing concern has arisen about adverse effects of this clinical practice on fetal brain development. We demonstrated previously that neonatal dexamethasone (DEX) treatment may render the newborn brain to be more vulnerable to hypoxia/ischemia (HI)-induced gray matter injury. Here, we examined whether neonatal DEX treatment may also affect the extent of HI-induced subcortical white matter (WM) injury in the developing rat brain. Using a HI model of premature brain injury, we demonstrated that a 3-day tapering course (0.5, 0.3, and 0.1 mg/kg) of DEX treatment in rat pups on postnatal days 1-3 (P1-3) significantly reduced the number of all stages of the oligodendroglial lineage cells on P7 and exacerbated HI-induced WM injury. Neonatal DEX treatment also enhanced HI-induced oligodendroglial apoptosis and astrocyte activation in the developing WM on P14. Likewise, HI-induced reductions in myelin thickness, axon caliber, and function during WM development were exacerbated by neonatal DEX treatment. Furthermore, neonatal DEX treatment further aggravated HI-induced motor deficits as assessed in the rotarod test. We also found that the administration of ß-lactam antibiotic ceftriaxone increased glutamate transporter-1 protein expression and significantly reduced HI-induced WM injury in neonatal DEX-treated rats. These results suggest that neonatal DEX treatment may lead the developing brain to be more vulnerable to subsequent HI-induced WM injury, which can be ameliorated by ceftriaxone administration.

Identifying and Managing Root Rot of Pulses on the Northern Great Plains.

Pulse crops (annual grain legumes such as field pea, lentil, dry bean, and chickpea) have become an important component of the cropping system in the northern Great Plains of North America over the last three decades. In many areas, the intensity of damping-off, seedling blight, root rot, and premature ripening of pulse crops is increasing, resulting in reduction in stand establishment and yield. This review provides a brief description of the important pathogens that make up the root rot complex and summarizes root rot management on pulses in the region. Initially, several specific Fusarium spp., a range of Pythium spp., and Rhizoctonia solani were identified as important components of the root rot disease complex. Molecular approaches have recently been used to identify the importance of Aphanomyces euteiches on pulses, and to demonstrate that year-to-year changes in precipitation and temperature have an important effect on pathogen prevalence. Progress has been made on management of root rot, but more IPM tools are required to provide effective disease management. Seed-treatment fungicides can reduce damping-off and seedling blight for many of the pathogens in this disease complex, but complex cocktails of active ingredients are required to protect seedlings from the pathogen complex present in most commercial fields. Partial resistance against many of the pathogens in the complex has been identified, but is not yet available in commercial cultivars. Cultural practices, especially diversified cropping rotations and early, shallow seeding, have been shown to have an important role in root rot management. Biocontrol agents may also have potential over the long term. Improved methods being developed to identify and quantify the pathogen inoculum in individual fields may help producers avoid high-risk fields and select IPM packages that enhance yield stability.

Sensitivity of Mycosphaerella pinodes to Pyraclostrobin Fungicide.

Mycosphaerella blight, caused by Mycosphaerella pinodes, is a destructive disease of field pea that is managed using foliar fungicides. Strobilurin fungicides have been used in western Canada for disease management since 2003. To assess the baseline sensitivities of M. pinodes isolates to the strobilurin fungicide pyraclostrobin, the effective concentration to reduce mycelial growth by 50% (EC50) was determined for 70 isolates collected prior to 2003 from Alberta, Saskatchewan, North Dakota, and Washington State. Each of these isolates was sensitive to pyraclostrobin, with EC50 values ranging from 0.03 to 0.29 mg liter-1. The pyraclostrobin concentrations required to reduce conidia germination by 50% was lower, ranging from 0.008 to 0.041 mg liter-1. In all, 324 isolates collected in 2010 and 2011 were tested for high levels of insensitivity by examining mycelial growth using a discriminatory dose of 5 mg liter-1. Nineteen isolates were highly insensitive to pyraclostrobin, with EC50 values of 80 to 216 mg liter-1. Conidia of these isolates germinated when exposed to a discriminatory dose of 0.1 mg liter-1. Insensitive isolates infected field pea plants treated with pyraclostrobin but sensitive isolates did not. The identification of insensitive isolates indicates that insensitivity may be emerging in the pathogen population.

Neonatal dexamethasone treatment exacerbates hypoxic-ischemic brain injury.

BACKGROUND: The synthetic glucocorticoid dexamethasone (DEX) is commonly used to prevent chronic lung disease in prematurely born infants. Treatment regimens usually consist of high doses of DEX for several weeks, notably during a critical period of brain development. Therefore, there is some concern about adverse effects of this clinical practice on fetal brain development. In this study, using a clinically relevant rat model, we examined the impact of neonatal DEX treatment on subsequent brain injury due to an episode of cerebral hypoxia-ischemia (HI). RESULTS: We found that a 3-day tapering course (0.5, 0.3 and 0.1 mg/kg) of DEX treatment in rat pups on postnatal days 1-3 (P1-3) exacerbated HI-induced brain injury on P7 by a glucocorticoid receptor-mediated mechanism. The aggravating effect of neonatal DEX treatment on HI-induced brain injury was correlated with decreased glutamate transporter-1 (GLT-1)-mediated glutamate reuptake. The expression levels of mRNA and protein of GLT-1 were significantly reduced by neonatal DEX treatment. We also found that the administration of ß-lactam antibiotic ceftriaxone increased GLT-1 protein expression and significantly reduced HI-induced brain injury in neonatal DEX-treated rats. CONCLUSIONS: These results suggest that early DEX exposure may lead the neonatal brain to be more vulnerable to subsequent HI injury, which can be ameliorated by administrating ceftriaxone.

A Specific and Sensitive Method for the Detection of Colletotrichum lindemuthianum in Dry Bean Tissue.

To facilitate early diagnosis and improve control of bean anthracnose, a rapid, specific, and sensitive polymerase chain reaction (PCR)-based method was developed to detect the causal agent, Colletotrichum lindemuthianum, in bean (Phaseolus vulgaris) seed. Based on sequence data of the rDNA region consisting of the 5.8S gene and internal transcribed spacers (ITS) 1 and 2 of four C. lindemuthianum races and 17 Colletotrichum species downloaded from GenBank, five forward primers were designed and evaluated for their specificity. Among them, one forward primer was selected for use in combination with ITS4 to specifically detect C. lindemuthianum. A 461-bp specific band was amplified from the genomic DNA template of 16 representative isolates of C. lindemuthianum, but not from 58 representative isolates of 17 other Colletotrichum species or 10 bean pathogens. Moreover, to enhance the sensitivity of detection, nested PCR was applied, which allowed the detection of as little as 10 fg of C. lindemuthianum genomic DNA and 1% infected seed powder, which was mixed with 99% healthy seed powder. The diagnostic analysis can be completed within 24 h, compared with about 2 weeks required for culturing. Furthermore, this method can be performed and interpreted by personnel with no specialized taxonomic expertise.

Detection and molecular characterization of an aster yellows phytoplasma in poker statice and Queen Anne's lace in Alberta, Canada.

Queen Anne's lace and poker statice plants were found with a yellows-type disease with typical phytoplasma symptoms in an experimental farm near Brooks, Alberta in 1996. Phytoplasma bodies were detected by transmission electron microscopy in phloem cells of symptomatic plants, but not in healthy plants. The presence of a phytoplasma was confirmed by analysis with the polymerase chain reaction. Using a pair of universal primer sequences derived from phytoplasma 16S rRNA, an amplified product of the expected size (1.2 kb) was observed in samples from infected plants, but not in asymptomatic plants. Sequence analysis of the PCR products from the 16S/23S rDNA intergenic spacer region indicated that the two phytoplasma isolates in Queen Anne's lace and poker statice are genetically closely related to the western aster yellows phytoplasma.