Sheet Images Research Articles

Image processing tools for petabyte-scale light sheet microscopy data.

Light sheet microscopy is a powerful technique for high-speed three-dimensional imaging of subcellular dynamics and large biological specimens. However, it often generates datasets ranging from hundreds of gigabytes to petabytes in size for a single experiment. Conventional computational tools process such images far slower than the time to acquire them and often fail outright due to memory limitations. To address these challenges, we present PetaKit5D, a scalable software solution for efficient petabyte-scale light sheet image processing. This software incorporates a suite of commonly used processing tools that are optimized for memory and performance. Notable advancements include rapid image readers and writers, fast and memory-efficient geometric transformations, high-performance Richardson-Lucy deconvolution and scalable Zarr-based stitching. These features outperform state-of-the-art methods by over one order of magnitude, enabling the processing of petabyte-scale image data at the full teravoxel rates of modern imaging cameras. The software opens new avenues for biological discoveries through large-scale imaging experiments.

Experimental Assessment of Paper Formation Conditions and Structural Two-Sidedness and Their Impacts on Curl Phenomena

Curl propensity is a critical-to-quality (CTQ) property of paper, as it causes severe problems during printing and other final conversion operations. The main papermaking factor causing the curl phenomenon is the existence of a fiber orientation (FO) gradient across the thickness direction (or ZD), also known as two-sidedness. Therefore, a methodology that characterizes the FO across the ZD is fundamental for papermakers. In this work, we propose and validate an efficient and cost-effective protocol based on sheet splitting and image analysis. Besides assessing the level of FO two-sidedness, the methodology also provides insights into the flow dynamics in the draining zone of the forming section of the paper machine and the drying stresses built into the paper. This information is relevant for monitoring, optimizing, and troubleshooting activities in the paper industry.

Deep Learning-Based Optical Music Recognition for Semantic Representation of Non-overlap and Overlap Music Notes

In the technology era, the process of teaching a computer to interpret musical notation is termed optical music recognition (OMR). It aims to convert musical note sheets presented in an image into a computer-readable format. Recently, the sequence-to-sequence model along with the attention mechanism (which is used in text and handwritten recognition) has been used in music notes recognition. However, due to the gradual disappearance of excessively long sequences of musical sheets, the mentioned OMR models which consist of long short-term memory are facing difficulties in learning the relationships among the musical notations. Consequently, a new framework has been proposed, leveraging the image segmentation technique to break up the procedure into several steps. In addition, an overlap problem in OMR has been addressed in this study. Overlapping can result in misinterpretation of music notations, producing inaccurate findings. Thus, a novel algorithm is being suggested to detect and segment the notations that are extremely close to each other. Our experiments are based on the usage of the Convolutional Neural Network block as a feature extractor from the image of the musical sheet and the sequence-to-sequence model to retrieve the corresponding semantic representation. The proposed approach is evaluated on The Printed Images of Music Staves dataset. The achieved results confirm that our suggested framework successfully solves the problem of long sequence music sheets, obtaining SER 0% for the non-overlap symbols in the best scenario. Furthermore, our approach has shown promising results in addressing the overlapping problem: 23.12 % SER for overlapping symbols.

TableStrRec: framework for table structure recognition in data sheet images

TableStrRec: framework for table structure recognition in data sheet images

A New Method for Calculating Reservoir Core-Bound Water Saturation Using the Cast Thin Section

The rock coring of the reservoir in the Bohai A field is difficult. The cores of the target section in the study area are loose, making it difficult to accurately measure the core-bound water saturation. The purpose of this research was to develop and validate a method for calculating a reservoir core-bound water saturation ratio using the cast thin section. First, pepper noise denoising and image enhancement were performed on the thin section by median filtering and gamma variation. Based on this, the enhanced sheet images were thresholded for segmentation by the two-dimensional OTSU algorithm, which automatically picked up the thin section pore-specific parameters. Then, the thin section image was equivalent to a capillary cross-section, while the thin film water fused to the pore surface was observed as bound water. For hydrophilic rocks with a strong homogeneity, the area of thin film water in the pore space of the sheet was divided by the total area of the pore space, which produced the bound water saturation. Next, the theoretical relationship between the film water thickness and the critical pore throat radius was derived based on the Young–Laplace equation. The bound water saturation of the rock was calculated by combining the pore perimeter and the area that was automatically picked up from the thin film for a given critical pore throat radius of the rock. Finally, 22 images of thin sections of sparse sandstone from the coring well section of the study area were image processed using the new method proposed in this paper, and the bound water saturation was calculated. The calculated results were compared with 22 NMR-bound water saturations and 11 semi-permeable baffle plate-bound water saturations in the same layer section. The results showed that the bound water saturation values calculated by the three methods produced consistent trends with absolute errors within 5%. The calculated results confirm the reliability of the method proposed in this paper. This method can effectively avoid the problem of the inaccurate results of core experiments due to the easy damage of sparse sandstone and provides a new idea for the accurate determination of the bound water saturation of sparse sandstone.

Impact of spray dynamics on combustion instabilities investigated by changing the atomizer recess in a swirl combustor

The present article is aimed at identifying the impact of spray-wall interaction on combustion instabilities. This is achieved by systematically modifying the liquid fuel atomizer location with respect to the injector outlet in experiments carried out on a single-injector swirling combustor. The injector is characterized by a high swirl number and a relatively high pressure drop, and features a pressure atomizer that delivers liquid heptane fuel in the form of a fine spray. In the first set of experiments, longitudinal self-sustained oscillations (SSOs) are examined by varying the atomizer recess distance and using three combustion chamber lengths. Three distinct instability regions characterized by jumps in amplitude and frequency are observed as a function of the atomizer recess. At lower SSO frequencies (obtained for longer chambers), the pressure fluctuation amplitude in the chamber decreases as the atomizer recess distance is reduced, tending towards a stable operation. An opposite behavior is observed at comparatively higher SSO frequencies (obtained with a shorter chamber), where the system becomes less unstable at higher atomizer recess distances. The oscillation frequency is also found to change and increases as the recess distance is reduced. Three recess distances corresponding to each zone of operation are selected to further analyze the flow and flame structures and dynamics. It is found that the mean velocity profiles only reveal moderate differences in the air flow field, but there are significant changes in the fuel spray distribution in the neighborhood of the injector axis. Laser sheet images of the spray reveal two patterns—one where the spray predominantly interacts with the injector end piece when the atomizer is recessed, and the other where this interaction is minimal. In the latter case, corresponding to a small recess, the fuel spray is directly conveyed into the chamber. It is next found that the spray distribution affects the flame pattern defining two distinct configurations. Finally, the flame response to external disturbances is characterized by flame describing functions (FDFs) that allow carrying out a stability analysis. FDF results obtained with two independent methods indicate that the gain and phase at the three recess positions vary substantially and that this can be used in combination with a low-order model to interpret the instability behavior of the system. This study might serve to guide the modeling of combustion dynamics and help design injectors that are less sensitive to instabilities.

Matching method for mutated veneer sheet images using gray-level co-occurrence matrix features

This paper studies the tracking of wooden veneer sheets by matching their respective wet and dry colour images. The tracking of veneer sheets has proved to be a challenging task due to random mutations during processing in terms of color changes, the emergence of defects, and, occasionally, lost pieces of the veneer surface. The proposed matching procedure involves image segmentation with five different sizes, followed by segment-wise extraction of Gray Level Co-occurrence Matrix (GLCM) textural feature arrays, and their similarity comparisons respectively. A voting mechanism is introduced that allocates the correct match based on the majority. An optional shifting procedure is applied to match candidates with missing areas. The method is demonstrated and benchmarked using a real-world dataset sourced from the industry, comprising 2579 high-quality images of spruce veneer pairs obtained from peeling and drying. In comparison to earlier studies that employed randomized 50 pair sampling on the same dataset, our approach yields a matching accuracy of 99.41%, outperforming the previously reported 84.93%. These findings have relevance for researchers in wood image analytics and carry practical implications for large-scale, automated veneer production facilities seeking innovative ways to optimize their raw material usage.

A deep learning model for predicting mechanical properties of polycrystalline graphene

Molecular dynamics simulations have been widely used to understand the properties of nanomaterials. The main issue associated with molecular dynamic modeling is the high computational costs. Machine learning (ML) models can be used as an alternative method for predicting the behavior of materials. After training, machine learning models can provide instantaneous results and avoid the computational costs of molecular dynamic simulations. We develop a deep convolutional neural network model to predict the mechanical properties of polycrystalline graphene. The data required for training our machine learning model is generated using molecular dynamics simulations by modeling the behavior of polycrystalline graphene under uniaxial tensile loading. More than 2000 data points are generated for graphene sheets of different grain sizes and grain orientations. The goal is to train the network such that it can predict the Young’s modulus and fracture stress of graphene sheets by analyzing an image of the polycrystalline sheet.

Senescence Reversion in Plant Images Using Perception and Unpaired Data

Recent work on using herbarium images for automatic plant identification, and in particular to do domain adaptation to field images, has been promising. A potential way to address such domain adaptation problem is generative: Hallucinate how a herbarium image would have looked like when it was in the field, and use such synthetic, fresh and green image for plant identification purposes. Such generative task, called senescence reversion, has been poorly explored. To our knowledge, it has been studied only in terms of paired data, meaning, with pairs of dry and fresh images of the same specimen. Such paired data is hard to produce, curate and find. In this work we explore herbarium senescence reversion via unpaired data. The lack of pairs at specimen level presents its own challenges, as capturing the intricacies of each species depends on images of different specimens from different domains. We explore learning a mapping from a herbarium image to leaf plant images and vice-versa, aligning two models, one for each herbarium and leaf domain. We experiment against the state-of-the-art paired baseline Pix2Pix which yields a SSIM of 0.8986, compared to our unpaired approach that yields 0.8865, showing very similar results on reconstruction metrics regardless of our approach not having the luxury of pairs by specimen. Additionally, we apply perceptual loss in order to improve the natural look of the synthetic images. The balance of how much perception is good to avoid reconstruction problems is also studied. Lastly, using a new unpaired dataset built by ourselves, our results show that using a low lambda value from 0.025 to 0.05 for perceptual loss, helps getting lower Frachet Inception Distances and higher Inception Scores. To our knowledge, no other work has focused on reverting senescence of herbarium sheet images based on unpaired data.

Method of railway shunting operation sheet information extraction guided by table header

The railway shunting operation sheet is the written plan of the shunting operation. In the current shunting operation preparation process, signalmen must input the plans manually for subsequent execution, which is inefficient and at risk of misoperation. Here, an information extraction method is proposed to convert shunting operation sheet images to readable instructions for Continuous Train Control (CTC) systems. First, the shunting operation sheet images are preprocessed to facilitate subsequent text recognition based on the Paddle OCR model. Second, the table is located and reconstructed with the results of recognized contents and coordinates of texts. Then, the results of the table reconstruction were optimized based on the tabulation rules of shunting operation. Finally, tabular texts containing structural and attribute information are transcribed into eXtensible Markup Language (XML) format, which can be read automatically by CTC systems based on the C programming language. The proposed approach was evaluated on real-world samples, and the results showed that the comprehensive accuracy of information extraction was 0.85. Furthermore, a pre-processing combination strategy was established to improve the accuracy and generalizability of the proposed method.

Production and characterization of Imperata cylindrica paper using potassium hydroxide as pulping agent

Abstract. Madung Z, Soloi S, Majid MHA, Sarjadi MS. 2022. Production and characterization of Imperata cylindrica paper using potassium hydroxide as pulping agent. Biodiversitas 23: 1490-1494. Non-woody plants have become an alternative fiber in paper production as woods, the primary sources of papers have become limited due to increasing demand of paper year by year. Imperata cylindrica or cogon grass as the source of fiber for the pulp and paper industry in replacing the wood fiber because it has high cellulose and low lignin content. In this study, cogon grasses were treated with various concentrations of potassium hydroxide, KOH (5%, 8%, 10% and 12%). The effects of alkali treatment on the properties of the papers were investigated in the present study. Scanning Electron Microscope (SEM) image of the paper sheet indicates that the amount of cellulose fiber and other materials were lessening toward a higher concentration of KOH treatment. This explained that most of the lignin was entirely removed and some cellulose probably degraded at higher KOH concentrations. Fourier Transform Infa-Red (FTIR) analysis showed that the peaks at range 1650 cm-1-1630 cm-1 attributed to the presence of v(C=C) aromatic ring of lignin disappeared at 10% and 12% of KOH. The mechanical strength of the papers decreased with an increasing concentration of KOH, while 5% of KOH produced the highest value of tensile strength.

Kinetics observations of bacterial cellulose thickness formation using image processing approach during the fermentation process

Bacterial cellulose is an exopolysaccharide that has a higher level of purity compared to cellulose from plants. Bacterial Cellulose (BC) is widely used for various uses so that it requires certain initial conditions, one of which is thickness. During the fermentation process, cellulose will be secreted into the medium to form BC sheets and visually visible over the time period. The aim of this research was to study the relationship between variables that influence during the fermentation and to fit the kinetic model of the BC thickness formation using image processing approach during the fermentation process. A USB camera was placed in front of the fermenter to capture the formation of BC thickness. Python programming language was used to process the image and calculated the thickness of the BC sheet from the beginning to the end of the fermentation process. Several supporting parameters were observed by placing the turbidity, pH, and medium temperature sensors. Observations were made in real time with a range of data collection every 15 mins during fermentation. The highest correlation value was obtained from the relationship between time and thickness. The fermentation process is divided into 2 clusters, a change in cluster occurs at the 61st hour. The model that describes the relationship between time and thickness was the Gompertz model.

The Herbarium 2021 Half-Earth Challenge Dataset and Machine Learning Competition.

Herbarium sheets present a unique view of the world's botanical history, evolution, and biodiversity. This makes them an all–important data source for botanical research. With the increased digitization of herbaria worldwide and advances in the domain of fine–grained visual classification which can facilitate automatic identification of herbarium specimen images, there are many opportunities for supporting and expanding research in this field. However, existing datasets are either too small, or not diverse enough, in terms of represented taxa, geographic distribution, and imaging protocols. Furthermore, aggregating datasets is difficult as taxa are recognized under a multitude of names and must be aligned to a common reference. We introduce the Herbarium 2021 Half–Earth dataset: the largest and most diverse dataset of herbarium specimen images, to date, for automatic taxon recognition. We also present the results of the Herbarium 2021 Half–Earth challenge, a competition that was part of the Eighth Workshop on Fine-Grained Visual Categorization (FGVC8) and hosted by Kaggle to encourage the development of models to automatically identify taxa from herbarium sheet images.

Real-Time Music Following in Score Sheet Images via Multi-Resolution Prediction

The task of real-time alignment between a music performance and the corresponding score (sheet music), also known as score following, poses a challenging multi-modal machine learning problem. Training a system that can solve this task robustly with live audio and real sheet music (i.e., scans or score images) requires precise ground truth alignments between audio and note-coordinate positions in the score sheet images. However, these kinds of annotations are difficult and costly to obtain, which is why research in this area mainly utilizes synthetic audio and sheet images to train and evaluate score following systems. In this work, we propose a method that does not solely rely on note alignments but is additionally capable of leveraging data with annotations of lower granularity, such as bar or score system alignments. This allows us to use a large collection of real-world piano performance recordings coarsely aligned to scanned score sheet images and, as a consequence, improve over current state-of-the-art approaches.

PEFCs Using Metallic Ti and Sn Electrocatalyst Supports

Introduction Currently, Pt-based catalyst on carbon black support (Pt/C) is widely used as the electrocatalyst for polymer electrolyte fuel cells (PEFCs). Carbon black support has high electronic conductivity and high specific surface area for supporting highly-dispersed Pt catalyst nanoparticles. However, under cathode condition of PEFCs e.g. during start-stop cycles, carbon black support can be electrochemically oxidized, causing detachment and aggregation of Pt catalyst nanoparticles (1). In order to solve this issue, we are developing SnO2 and TiO2 supports, which are relatively stable even under cathode conditions, coated on carbon-based materials as the conductive backbone, exhibiting high durability (2, 3). However, as far as we use carbon materials, carbon corrosion problem may still be remained. Here in this study, porous metallic Ti and Sn sheets are applied as catalyst support. Such electrodes may exhibit high durability under PEFC cathode conditions due to the formation of stable TiO2 and SnO2 layers on the outermost surface of Ti and Sn metals. Furthermore, metallic sheets with porous structures also have a potential to act as gas diffusion layers (GDL). This study aims to prepare such electrodes integrating electrocatalysts with high conductivity and high durability. Experimental For preparing metallic Ti supported electrodes, porous Ti sintered sheets were used. First, the porous Ti sintered sheets were chemically etched in NaOHaq to increase their surface area. The etched Ti sheets were then heat-treated at 400 °C in 5%H2-N2. Pt catalyst nanoparticles were then decorated on the Ti sheets by the arc plasma deposition (APD) or the acac method (4). We also prepared Nb-doped Ti sheets to improve their electronic conductivity. For preparing metallic Sn supported electrodes, cellulose-based filter paper was used as a template. First, SnO2 sol was impregnated into the filter papers in vacuum, followed by heat treatment in oxidizing atmosphere to remove the cellulose-based templates. After that, porous metallic Sn sheets were obtained by reducing heat-treatment. The Pt catalyst nanoparticles were decorated on the porous Sn sheets by the acac method (4). Microstructures of the porous Pt/Ti and Pt/Sn sheets were observed by FE-SEM and STEM-EDS. Their electrochemical activities were measured in a half-cell setup in HClO4 electrolyte. Results and Discussion Figure 1 shows an FE-SEM image of the porous Ti sheet after the surface treatment by NaOHaq. The surface of the Ti sheets after the surface treatment has needle-like TiO2 nanostructures with a diameter of several nm, with high surface area. Figure 2 shows the linear sweep voltammograms by the half-cell tests for the Pt/Ti sheet electrodes. The Pt/Ti sheet electrode prepared by the acac method may have higher contact resistance than the Pt/Ti sheet electrode by the APD method. However, such contact resistance may be reduced by Nb-doping the Ti sheets. Figure 3 shows an FE-SEM image of the prepared Sn sheet for catalyst support. We could prepare metallic Sn sheets with porous structures derived from the filter paper template. However, optimization of heat-treatment conditions is further required. It is because the reduction of SnO2 is difficult if the heat-treatment temperature is too low, but Sn will be melted if the temperature is too high. Acknowledgement Financial support from New Energy and Industrial Technology Development Organization (NEDO) is gratefully acknowledged (Contract No. 20001214-0). References L. M. Roen, C. H. Pail, and T. D. Jarvi, Electrochem. Solid-State Lett., 7 (1), A19 (2004). S. Matsumoto, M. Nagamine, Z. Noda, J. Matsuda, S. M. Lyth, A. Hayashi, and K. Sasaki, J. Electrochem. Soc., 165 (14), 1165 (2018). Y. Nakazato, D. Kawachino, Z. Noda, J. Matsuda, A. Hayashi, and K. Sasaki, ECS Trans., 80 (8), 897 (2017).A. Hayashi, H. Notsu, K. Kijima, J. Miyamoto, and I. Yagi, Electrochim. Acta, 53 (21), 6117 (2008). Figure 1

TableDet: An end-to-end deep learning approach for table detection and table image classification in data sheet images

Global supply chains are kept viable through the information shared through billions of electronic documents, many of which extensively use tables to display critical information. Making effective supply chain decisions requires the extraction of data from these tables which is hindered by the variations in layouts and styles of tables. In this paper, we propose Table Det: a deep learning based methodology to solve table detection and table image classification in data sheet images in a single inference as the first stage of the table text extraction pipeline. TableDet utilizes Cascade R-CNN with Complete IOU (CIOU) loss and a deformable convolution backbone as its underlying architecture to capture the variations in scales and orientations of tables. It also detects text and figures to enhance its table detection performance. We demonstrate the effectiveness of training TableDet with a dual-step transfer learning process and fine-tuning it with Table Aware Cutout (TAC) augmentation strategy. We achieved the highest F1 score for table detection against state-of-the-art solutions on ICDAR 2013 (complete set), ICDAR 2017 (test set) and ICDAR 2019 (test set) with 100%, 99.3% and 95.1% respectively. For the table image classification task we attained 100% recall and above 85% precision on three test sets. This classification capability ensures that all images with tables would be promoted to the next step in the table text extraction pipeline, with a small number of images without tables making it through.

Robust Machine Learning Classification of Unlabeled Biological Data: A case study with herbaria sheets

There is much excitement across a broad range of biological disciplines over the prospect of using deep learning and similar modern statistical methods to label research data. The extensive time, effort, and cost required for humans to label a dataset drastically limits the type and amount of data that can be reasonably utilized, and is currently a major bottleneck to the extensive application of biological datasets such as specimen imagery, video and audio recordings. While a number of researchers have shown how deep convolutional neural networks (CNN) can be trained to classify image data with 80-90% accuracy, that range of accuracy is still too low for most research applications. Furthermore, applying these classifiers to new, unlabeled data from a dataset other than the one used for training the classifier would likely result in even lower accuracy. As a result, these classifiers have still not generally been applied to unlabeled data—which is where they could be most useful. In this talk, we will present a method for determining a confidence metric on predicted classifications (i.e. "labels") from a deep CNN classifier that can inform a user whether to trust a particular automatic label or to discard it, thereby giving a reasonable and straight-forward method to label a previously unlabeled dataset with high confidence. Essentially, it is an approach that allows an imperfect method of classification to be used in a useful way that can save an enormous amount of time and effort and/or greatly increase the amount of data that can be reasonably utilized. In this work, the training dataset consisted of a set of records of flowering plant species that collectively exhibited a range of reproductive morphologies, represented multiple taxonomic groups, and could be easily scored by humans for reproductive condition by examination of specimen images. The records were labeled as reproductive, budding, flowering and/or fruiting. All of the data and images were obtained from the Consortium of Northeastern Herbaria portal (CNH). There were two unscored datasets that were used to evaluate the classifiers. One dataset contained the same taxa that were in the training dataset and the second dataset contained all remaining flowering plant taxa in the CNH portal database that were not included in the other two datasets. Records of families with flowers that are obscure (i.e., they lack petals & sepals or have vestigial structures) were excluded. To label the reproductive state of the plants, we trained one deep CNN classifier using the XCeption architecture for the binary classification of each state (e.g., budding vs. not budding). This method and architecture was chosen because of its success in similar image-classification tasks. Each of these networks takes an image of a herbarium sheet as input, and outputs a value in the interval [0,1]. In these networks, the output is typically thresholded to generate a binary label, but we found it could also be used to approximate a measure of confidence in the network’s classification. By treating this value as a confidence metric, we are able to input a large unlabeled dataset into the classifier and then trust the labels that were assigned a “high confidence” and leave the remainder unlabeled. After training the network, the performance of the four classifiers (reproductive, budding, flowering, fruiting) achieved 85-90% accuracy compared to expert-labeled data. However, as described above, the real value of these approaches comes from their prospects for labelling previously unlabeled data, thus helping to replace expensive and time-consuming human labor. We then applied our confidence-interval-based approach to a collection of 600k images and were able to label 35-70% of the samples with a chosen confidence threshold of 95%. In other words, we could then use the high-confidence labels and simply not automatically label the remaining unclassifiable samples. The data from these samples could then be labeled manually, or, if appropriate, not labeled at all.

Automated Extraction of Phenotypic Leaf Traits of Individual Intact Herbarium Leaves from Herbarium Specimen Images Using Deep Learning Based Semantic Segmentation.

With the increase in the digitization efforts of herbarium collections worldwide, dataset repositories such as iDigBio and GBIF now have hundreds of thousands of herbarium sheet images ready for exploration. Although this serves as a new source of plant leaves data, herbarium datasets have an inherent challenge to deal with the sheets containing other non-plant objects such as color charts, barcodes, and labels. Even for the plant part itself, a combination of different overlapping, damaged, and intact individual leaves exist together with other plant organs such as stems and fruits, which increases the complexity of leaf trait extraction and analysis. Focusing on segmentation and trait extraction on individual intact herbarium leaves, this study proposes a pipeline consisting of deep learning semantic segmentation model (DeepLabv3+), connected component analysis, and a single-leaf classifier trained on binary images to automate the extraction of an intact individual leaf with phenotypic traits. The proposed method achieved a higher F1-score for both the in-house dataset (96%) and on a publicly available herbarium dataset (93%) compared to object detection-based approaches including Faster R-CNN and YOLOv5. Furthermore, using the proposed approach, the phenotypic measurements extracted from the segmented individual leaves were closer to the ground truth measurements, which suggests the importance of the segmentation process in handling background noise. Compared to the object detection-based approaches, the proposed method showed a promising direction toward an autonomous tool for the extraction of individual leaves together with their trait data directly from herbarium specimen images.

Designing the Cathode in MEA Using Mesoporous Carbon Fibers

Introduction PEFCs have been known as one of environment-friendly power sources and been used for both mobile and stationary applications. However, the high cost, which mostly comes from the fabrication of MEAs made by carbon, Pt and polymer electrolytes, has been a great wall on the way of wide commercialization of PEFCs. For electrocatalysts, which mostly determine PEFC performance, the better performance and high durability with low noble metal usage are highly desired. In our research group, mesoporous carbon (MC) materials have been studied as catalyst supports, and encapsulation of Pt into controlled nm-space has led to improvement in durability [1]. However, we have found a problem in the disordered μm-structure. Therefore, we are now focusing on developing the fiberized MC, mesoporous carbon fibers (MCFs), where both the nm- and μm-structures are controllable. In our previous study, fundamental electrochemical properties of MCFs powder catalysts were evaluated with half-cell measurements in solution [2]. In this study, MCF sheets were newly used as catalyst supports. MEAs with Pt deposited MCF sheet cathode and related cathodes were made and electrochemically evaluated . Experimental MCFs with mesoporous diameter of 6-7 nm were synthesized according to our previous study [2], which was based on combining the self-assembly of the surfactant-carbon precursor and an electrospinning deposition process. MCFs sheets were cut into 1 x 1 cm2, but a part of them were further ball-milled into the μm-scale powder. After the Pt deposition on MCFs described in our previous study [1], three different cathodes, Pt deposited milled MCFs, TKK Pt/KB (TEC10E50E), and with TKK Pt/KB were prepared by keeping the total Pt loading as 0.3 mg/cm2. MCF sheet with TKK Pt/KB was built by spray printing a Pt/KB layer on the Nafion212 first and then on the MCFs sheet, which was placed on the first layer of Pt/KB. Results and discussion A cross-sectional image of MCFs sheet was taken using a FIB fabrication method and shown in Fig.1. The thickness of a single layer of MCFs was 1.0 to 1.3 μm. In order to keep the total weight of carbon close to Pt/milled MCF cathode, two layers of MCFs were used as a sheet in this study. IV performance, corresponding power density, and ohmic overvoltage of MEAs with different cathodes are shown in Fig. 2. The current density at 0.8 V in IV curves was However, lower ohmic overvoltage was observed in MEA made by MCFs sheet with TKK Pt/KB. The obtained resistance was about 70 mΩ, which was 20 mΩ less and 30 mΩ less comparing to MEA with Pt/KB and MEA with Pt/milled MCFs, respectively. We have found that the usage of MCF sheets leads to thin electrodes, resulting in reduction in ohmic resistance. In order to further improve the activation and concentration overvoltage, Pt deposited MCFs sheet with Pt/KB cathode is under the study and will be reported.

High performance area measurement system for leather industries

India is one of the leading producers of leather and leather products. Leather Sheet provided by the Tanners is in irregular shape and one of the main factors of its cost is based on its surface area. To calculate the size of the leather, different methods are used which include manual, mechanical and electronics systems. The presently used systems are offer greater amount of error in the measurement due to its irregular shape and cost of the system also high. The proposed research work is development of a prototype with surface area measurement system based on image processing in which a digital photographic image of leather sheet is placed on a conveyer belt having Variable Frequency Drive controlled rollers. IR sensors are used as feedback mechanism for practical implementation of the system which is used to minimize the error in surface area calculations. The main objective is to develop a flat surface area measuring system which is used to calculate the surface area of any irregular sheet. The irregular leather sheet is used in this work. The system is self-protected by user name and password set through software for security purpose. Only authorized user can enter into the system by entering the valid pin code. After entering into the system, the user can measure the area of any irregular sheet, monitor and control the system. The heart of the system is Microcontroller (Atmega328p) which controls the complete working of the system. The controlling instructions for the system are given through the designed Human to Machine Interface (HMI). For communication purpose the GSM modem is also interfaced with the Arduino Microcontroller. The remote user can also monitor the current status of the devices by sending SMS message to the GSM modem.