Search Results (10)

Micro-crack defects on the surfaces of plate heat exchanger sheets often exhibit a linear grayscale pattern when clustered. In defect detection, traditional methods are more suitable than deep learning models in controlled production environments with limited computing resources to meet stringent national standards, which require low miss rates. However, deep learning models commonly suffer feature loss when detecting individual, small-scale defects, leading to higher leak detection rates. Moreover, in grayscale image line detection using traditional methods, the varying direction, width, and asymmetric grayscale profiles of defects can result in filled grayscale valleys due to width-adaptive smoothing coefficients, complicating accurate defect extraction. To address these issues, this study establishes a theoretical foundation for parameter selection in variable-width defect detection. We propose a directional gradient-based algorithm that mathematically constrains the Gaussian template width to cover variable-width defects with a fixed σ, reframing the detection defect from ridge edges to centrally symmetric double-ridge edges in gradient images. Experimental results show that, when tested in the defective boards library and under simulated factory CPU conditions, this algorithm achieves a miss detection rate of 14.55%, a false detection rate of 21.85%, and an 600 × 600 pixel image detection time of 0.1402 s. Compared to traditional line detection and deep learning object detection methods, this algorithm proves advantageous for detecting micro-crack defects on plate heat exchanger sheets in industrial production, particularly in data-scarce and resource-limited scenarios. Full article

Accurately identifying individual wildlife is critical to effective species management and conservation efforts. However, it becomes particularly challenging when distinctive features, such as spot shape and size, serve as primary discriminators, as in the case of Sika deer. To address this challenge, we employed four different Convolutional Neural Network (CNN) base models (EfficientNetB7, VGG19, ResNet152, Inception_v3) within a Siamese Network Architecture that used triplet loss functions for the identification and re-identification of Sika deer. Subsequently, we then determined the best-performing model based on its ability to capture discriminative features. From this model, we extracted embeddings representing the learned features. We then applied a Support Vector Machine (SVM) to these embeddings to classify individual Sika deer. We analyzed 5169 image datasets consisting of images of seven individual Sika deers captured with three camera traps deployed on farmland in Hokkaido, Japan, for over 60 days. During our analysis, ResNet152 performed exceptionally well, achieving a training accuracy of 0.97, and a validation accuracy of 0.96, with mAP scores for the training and validation datasets of 0.97 and 0.96, respectively. We extracted 128 dimensional embeddings of ResNet152 and performed Principal Component Analysis (PCA) for dimensionality reduction. PCA1 and PCA2, which together accounted for over 80% of the variance collectively, were selected for subsequent SVM analysis. Utilizing the Radial Basis Function (RBF) kernel, which yielded a cross-validation score of 0.96, proved to be most suitable for our research. Hyperparameter optimization using the GridSearchCV library resulted in a gamma value of 10 and C value of 0.001. The OneVsRest SVM classifier achieved an impressive overall accuracy of 0.97 and 0.96, respectively, for the training and validation datasets. This study presents a precise model for identifying individual Sika deer using images and video frames, which can be replicated for other species with unique patterns, thereby assisting conservationists and researchers in effectively monitoring and protecting the species. Full article

Deep neural networks (DNNs) dominate many tasks in the computer vision domain, but it is still difficult to understand and interpret the information contained within these networks. To gain better insight into how a network learns and operates, there is a strong need to visualize these complex structures, and this remains an important research direction. In this paper, we address the problem of how the interactive display of DNNs in a virtual reality (VR) setup can be used for general understanding and architectural assessment. We compiled a static library as a plugin for the Caffe framework in the Unity gaming engine. We used routines from this plugin to create and visualize a VR-based AlexNet architecture for an image classification task. Our layered interactive model allows the user to freely navigate back and forth within the network during visual exploration. To make the DNN model even more accessible, the user can select certain connections to understand the activity flow at a particular neuron. Our VR setup also allows users to hide the activation maps/filters or even interactively occlude certain features in an image in real-time. Furthermore, we added an interpretation module and reframed the Shapley values to give a deeper understanding of the different layers. Thus, this novel tool offers more direct access to network structures and results, and its immersive operation is especially instructive for both novices and experts in the field of DNNs. Full article

Mount components and textile borders represent important elements of Asian paintings. However, they are often side-lined or not considered an integral part of the original piece, as they may be later additions or may have been replaced during historic conservation or mounting interventions. Nevertheless, evidence is sometimes present that textile borders are contemporaneous to the production of the paintings they frame or, in the case of paintings found in archaeological contexts, to the time of deposition. Even when not contemporaneous with the paintings, the mount textiles are often of significant historic interest in themselves, showing a range of complex textile techniques and materials, and highlighting the re-use of fabrics. In all these cases, the study and reconstruction of the original colours of the borders enable further understanding of the holistic visual impact originally intended for the composition, as well as of the role of colour itself, which was used to emphasise, complement or contrast important pictorial themes or motifs in the paintings. Furthermore, the identification of dyes and dyeing techniques has the potential to support the production date and provenance of the paintings. In this study, the textile borders and some additional mounting elements of six paintings (late 9th–10th century CE) from the Library Cave, Mogao Grottoes, Dunhuang, China, one rare Korean portrait painting dated 1789 CE, and two Tibetan thangkas (18th century) were investigated with the aim to identify the dyes present. Fibre optic reflectance spectroscopy (FORS) was used to obtain information non-invasively and, when sampling was possible, high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) was used to obtain molecular identification of the dyestuffs employed in their production. Typical Asian dyes, such as gromwell (Lithospermum erythrorhizon), sappanwood (Biancaea sappan), safflower (Carthamus tinctorius), turmeric (Curcuma longa) and pagoda tree flower buds (Sophora japonica), were identified. Some of the dyeing techniques were commensurate with the geographical and temporal provenance assigned to these pieces. Considerations about fading and discolouration of the dyes enabled valuable additional information to be obtained that complements the evidence gleaned from the study of the paintings and informs conservators and curators on best practices in the preservation and display of these precious and delicate artworks. Full article

Zika virus (ZIKV) was identified in 1947 in the Zika forest of Uganda and it has emerged recently as a global health threat, with recurring outbreaks and its associations with congenital microcephaly through maternal fetal transmission and Guillain-Barré syndrome. Currently, there are no United States (US) Food and Drug Administration (FDA)-approved vaccines or antivirals to treat ZIKV infections, which underscores an urgent medical need for the development of disease intervention strategies to treat ZIKV infection and associated disease. Drug repurposing offers various advantages over developing an entirely new drug by significantly reducing the timeline and resources required to advance a candidate antiviral into the clinic. Screening the ReFRAME library, we identified ten compounds with antiviral activity against the prototypic mammarenavirus lymphocytic choriomeningitis virus (LCMV). Moreover, we showed the ability of these ten compounds to inhibit influenza A and B virus infections, supporting their broad-spectrum antiviral activity. In this study, we further evaluated the broad-spectrum antiviral activity of the ten identified compounds by testing their activity against ZIKV. Among the ten compounds, Azaribine (SI-MTT = 146.29), AVN-944 (SI-MTT = 278.16), and Brequinar (SI-MTT = 157.42) showed potent anti-ZIKV activity in post-treatment therapeutic conditions. We also observed potent anti-ZIKV activity for Mycophenolate mofetil (SI-MTT = 20.51), Mycophenolic acid (SI-MTT = 36.33), and AVN-944 (SI-MTT = 24.51) in pre-treatment prophylactic conditions and potent co-treatment inhibitory activity for Obatoclax (SI-MTT = 60.58), Azaribine (SI-MTT = 91.51), and Mycophenolate mofetil (SI-MTT = 73.26) in co-treatment conditions. Importantly, the inhibitory effect of these compounds was strain independent, as they similarly inhibited ZIKV strains from both African and Asian/American lineages. Our results support the broad-spectrum antiviral activity of these ten compounds and suggest their use for the development of antiviral treatment options of ZIKV infection. Full article

Chagas disease, caused by the kinetoplastid parasite Trypanosoma cruzi, affects between 6 and 7 million people worldwide, with an estimated 300,000 to 1 million of these cases in the United States. In the chronic phase of infection, T. cruzi can cause severe gastrointestinal and cardiac disease, which can be fatal. Currently, only benznidazole is clinically approved by the FDA for pediatric use to treat this infection in the USA. Toxicity associated with this compound has driven the search for new anti-Chagas agents. Drug repurposing is a particularly attractive strategy for neglected diseases, as pharmacological parameters and toxicity are already known for these compounds, reducing costs and saving time in the drug development pipeline. Here, we screened 7680 compounds from the Repurposing, Focused Rescue, and Accelerated Medchem (ReFRAME) library, a collection of drugs or compounds with confirmed clinical safety, against T. cruzi. We identified seven compounds of interest with potent in vitro activity against the parasite with a therapeutic index of 10 or greater, including the previously unreported activity of the antiherpetic compound 348U87. These results provide the framework for further development of new T. cruzi leads that can potentially move quickly to the clinic. Full article

Soil organic carbon (SOC) loss is one of the main causes of soil degradation in croplands. Thus, spatial and temporal monitoring of SOC is extremely important, both from the environmental and economic perspective. In this regard, the high temporal, spatial, and spectral resolution of the Sentinel-2 data can be exploited for monitoring SOC contents in the topsoil of croplands. In this study, we aim to test the effect of the threshold for a spectral index linked to soil moisture and crop residues on the performance of SOC prediction models using the Multi-Spectral Instrument (MSI) Sentinel-2 and the European Land Use/cover Area frame Statistical survey (LUCAS) topsoil database. The LUCAS spectral data resampled according to MSI/Sentinel-2 bands, which were used to build SOC prediction models combining pairs of the bands. The SOC models were applied to a Sentinel-2 image acquired in North-Eastern Germany after removing the pixels characterized by clouds and green vegetation. Then, we tested different thresholds of the Normalized Burn Ratio 2 (NBR2) index in order to mask moist soil pixels and those with dry vegetation and crop residues. The model accuracy was tested on an independent validation database and the best ratio of performance to deviation (RPD) was obtained using the average between bands B6 and B5 (Red-Edge Carbon Index: RE-CI) (RPD: 4.4) and between B4 and B5 (Red-Red-Edge Carbon Index: RRE-CI) (RPD: 2.9) for a very low NBR2 threshold (0.05). Employing a higher NBR2 tolerance (higher NBR2 values), the mapped area increases to the detriment of the validation accuracy. The proposed approach allowed us to accurately map SOC over a large area exploiting the LUCAS spectral library and, thus, avoid a new ad hoc field campaign. Moreover, the threshold for selecting the bare soil pixels can be tuned, according to the goal of the survey. The quality of the SOC map for each tolerance level can be judged based on the figures of merit of the model. Full article

Phytocystatins play multiple roles in plant growth, development and resistance to pests and other environmental stresses. A ramie (Boehmeria nivea L.) phytocystatin gene, designated as BnCPI, was isolated from a ramie cDNA library and its full-length cDNA was obtained by rapid amplification of cDNA ends (RACE). The full-length cDNA sequence (691 bp) consisted of a 303 bp open reading frame (ORF) encoding a protein of 100 amino acids with deduced molecular mass of 11.06 kDa and a theoretical isoelectric point (pI) of 6.0. The alignment of genome DNA (accession no. MF153097) and cDNA sequences of BnCPI showed that an intron (~104 bp) exists in the coding region. The BnCPI protein contains most of the highly conserved blocks including Gly⁵-Gly⁶ at the N-terminal, the reactive site motif QxVxG (Q⁴⁹V⁵⁰V⁵¹S⁵²G⁵³), the L⁷⁹-W⁸⁰ block and the [LVI]-[AGT]-[RKE]-[FY]-[AS]-[VI]-x-[EDQV]-[HYFQ]-N (L²²G²³R²⁴ F²⁵A²⁶V²⁷ D²⁸D²⁹H³⁰ N³¹) block that is common among plant cystatins. BLAST analysis indicated that BnCPI is similar to cystatins from Glycine max (77%), Glycine soja (76%), Hevea brasiliensis (75%) and Ricinus communis (75%). The BnCPI was subcloned into expression vector pSmart-I and then overexpressed in Escherichia coli BL21 (DE3) as a His-tagged recombinant protein. The purified reBnCPI has a molecular mass of 11.4 kDa determined by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE). Purified reBnCPI can efficiently inhibit the protease activity of papain and ficin toward BANA (Nα-benzoyl-L-arginine-2-naphthyamide), as well as the mycelium growth of some important plant pathogenic fungi. The data further contribute to our understanding of the molecular functions of BnCPI. Full article

The extraordinary scientific-technical, economic, and social transformations related to the widespread use of computers and to the whole information and communication technologies have not been accompanied by the development of a scientific “informational” perspective helping make a coherent sense of the spectacular changes occurring. Like in other industrial revolutions of the past, technical praxis antedates the emergence of theoretical disciplines. Apart from the difficulties in handling new empirical domains and in framing new ways of thinking, the case of information science implies the difficult re-evaluation of important bodies of knowledge already well accommodated in specific disciplines. Herein, we will discuss how a new understanding of the “natural information flows” as they prototypically occur in living beings—even in the simplest cells—could provide a sound basis for reappraising fundamental problems of the new science. The role of a renewed information science, multidisciplinarily conceived and empirically grounded, widely transcends the limited “library” and knowledge-repositories mission into which classical information science was cajoled during past decades. Paraphrasing the Spanish philosopher J. Ortega y Gasset, the overhaul of information science itself becomes “the challenge of our time”. Full article

Substrate identification is the key to defining molecular pathways or cellular processes regulated by proteases. Although phage display with random peptide libraries has been used to analyze substrate specificity of proteases, it is difficult to deduce endogenous substrates from mapped peptide motifs. Phage display with conventional cDNA libraries identifies high percentage of non-open reading frame (non-ORF) clones, which encode short unnatural peptides, owing to uncontrollable reading frames of cellular proteins. We recently developed ORF phage display to identify endogenous proteins with specific binding or functional activity with minimal reading frame problem. Here we used calpain 2 as a protease to demonstrate that ORF phage display is capable of identifying endogenous substrates and showed its advantage to re-verify and characterize the identified substrates without requiring pure substrate proteins. An ORF phage display cDNA library with C-terminal biotin was bound to immobilized streptavidin and released by cleavage with calpain 2. After three rounds of phage selection, eleven substrates were identified, including calpastatin of endogenous calpain inhibitor. These results suggest that ORF phage display is a valuable technology to identify endogenous substrates for proteases. Full article