Vertical Pool Research Articles

Experimental investigation on a new method for modifying the geometry of a vertical drop edge equipped with a screen on the amount of energy dissipation

Abstract The aim of the current research is to investigate the effect of geometry and different angles (180°, 150°, 120° and 90°) of the edge of a vertical drop with a height of 0.15 m with and without a vertical screen on energy dissipation. The finding showed that the triangulation plane forms a vertical drop at different angles of 180°, 150°, 120° and 90° without a screen that increases the energy dissipation of the drop, which is highest at an angle of 90°, by 50.5%. In the vertical drop with a screen, reducing the angle also leads to a decrease in energy dissipation, so the highest energy dissipation occurs at an angle of 180° with a 45.2% increase in energy dissipation compared to the vertical drop without a screen. The relative depth of the edge and the relative depth of the vertical drop pool of the triangular plane are different at the angle vertex and its wings, in such a way that the relative depth of the edge at the vertex is less than the wings and the relative depth of the pool at the vertex is greater than the wings.

A dense multi‐scale context and asymmetric pooling embedding network for smoke segmentation

AbstractIt is very challenging to accurately segment smoke images because smoke has some adverse vision characteristics, such as anomalous shapes, blurred edges, and translucency. Existing methods cannot fully focus on the texture details of anomalous shapes and blurred edges simultaneously. To solve these problems, a Dense Multi‐scale context and Asymmetric pooling Embedding Network (DMAENet) is proposed to model the smoke edge details and anomalous shapes for smoke segmentation. To capture the feature information from different scales, a Dense Multi‐scale Context Module (DMCM) is proposed to further enhance the feature representation capability of our network under the help of asymmetric convolutions. To efficiently extract features for long‐shaped objects, the authors use asymmetric pooling to propose an Asymmetric Pooling Enhancement Module (APEM). The vertical and horizontal pooling methods are responsible for enhancing features of irregular objects. Finally, a Feature Fusion Module (FFM) is designed, which accepts three inputs for improving performance. Low and high‐level features are fused by pixel‐wise summing, and then the summed feature maps are further enhanced in an attention manner. Experimental results on synthetic and real smoke datasets validate that all these modules can improve performance, and the proposed DMAENet obviously outperforms existing state‐of‐the‐art methods.

Predicting Big Babies: Using Amniotic Fluid and Baby Weight Estimates

Background: Macrosomia or big baby is defined as a newborn with a birth weight of more than 4kg. Amniotic fluid index (AFI) which is estimated by adding length of deepest vertical pool in four corners of liquor in uterus is an indicator of fetal well-being. This parameter when combined with ultrasonographic estimated fetal weight can predict macrosomia accurately. Objective: To evaluate diagnostic accuracy of integrated examination with amniotic fluid index (AFI) and estimated fetal weight (EFW) for prediction of macrosomia. Study Design: Cross sectional study. Settings: Department of Obstetrics and Gynecology, Combined Military Hospital (CMH), Rawalpindi Pakistan. Duration: 6 months from 29th October 2019 to 28th April 2020. Methods: After meeting the inclusion criteria 250 females were enrolled. All patients underwent sonographic evaluation for AFI and EFW estimation. Patients were categorized as positive or negative for macrosomia. Then all females were followed-up in OPD till delivery. If birth weight was ≥4000 grams, then macrosomia was confirmed. Results: The average age of the females was 27.21 ± 4.59 years, there were 55(22%) females as nulliparous. The sensitivity, specificity, and diagnostic accuracy of integrated examination of AFI & EFW were 50%, 98.78%, & 98% for definition of macrosomia as base line. Conclusion: This study showed good diagnostic accuracy of integrated amniotic fluid index and ultrasound baby weight estimates as predictor of macrosomia having low sensitivity.

Flame geometric characteristics of large-scale pool fires under controlled wind conditions

This paper presents an experimental study on the flame geometric characteristics of large-scale open pool fires under well-controlled wind conditions. Square fuel pools with combustion areas of 50 m2, 100 m2, 200 m2, 380 m2, and 400 m2 were used to burn RP-5 aviation fuel and diesel oil. A large-scale open jet wind tunnel generated a well-controlled and regulated horizontal wind field with wind speed (U) of up to 17 m/s. It was found that the evolution of large-scale pool fires under wind conditions involved four regimes, namely free vertical pool fire, tilted pool fire without flame drag, tilted pool fire with stable flame drag, and tilted pool fire close to the ground. The wind significantly affected the flame shape, the vortex structures, and the smoke layer characteristics around the flame. Five quantities of flame geometry were measured, including the flame base drag (ld), the flame length (L), the vertical flame height (H), and two flame tilt angles of θ1 (defined by the fuel pool center) and θ2 (defined by the flame base center). The critical wind speeds for flame base drag were obtained for the first time, and it showed a steady increase with the fuel pool size. With increasing U, the vertical flame height (H) decreased steadily, while the flame length (L) increased slightly, and the flame base drag (ld) and the flame tilt angles (θ1, θ2) increased steadily and leveled off for U>7–8 m/s. By introducing the buoyancy length scale and integrating the mixture fraction equation, the correlations for the five flame geometric characteristics were derived, which agreed well with the data of large-scale pool fires in this work and literature. The correlation for flame base drag also successfully predicted the critical wind speed for the occurrence of the flame drag.

Accurate Retinal Vessel Segmentation in Color Fundus Images via Fully Attention-Based Networks.

Automatic retinal vessel segmentation is important for the diagnosis and prevention of ophthalmic diseases. The existing deep learning retinal vessel segmentation models always treat each pixel equally. However, the multi-scale vessel structure is a vital factor affecting the segmentation results, especially in thin vessels. To address this crucial gap, we propose a novel Fully Attention-based Network (FANet) based on attention mechanisms to adaptively learn rich feature representation and aggregate the multi-scale information. Specifically, the framework consists of the image pre-processing procedure and the semantic segmentation networks. Green channel extraction (GE) and contrast limited adaptive histogram equalization (CLAHE) are employed as pre-processing to enhance the texture and contrast of retinal blood images. Besides, the network combines two types of attention modules with the U-Net. We propose a lightweight dual-direction attention block to model global dependencies and reduce intra-class inconsistencies, in which the weights of feature maps are updated based on the semantic correlation between pixels. The dual-direction attention block utilizes horizontal and vertical pooling operations to produce the attention map. In this way, the network aggregates global contextual information from semantic-closer regions or a series of pixels belonging to the same object category. Meanwhile, we adopt the selective kernel (SK) unit to replace the standard convolution for obtaining multi-scale features of different receptive field sizes generated by soft attention. Furthermore, we demonstrate that the proposed model can effectively identify irregular, noisy, and multi-scale retinal vessels. The abundant experiments on DRIVE, STARE, and CHASE_DB1 datasets show that our method achieves state-of-the-art performance.

Estimation of system code SOCRAT/V3 accuracy to simulate the heat transfer in a pool of volumetrically heated liquid on the basis of BAFOND experiments

The accuracy of system code SOCRAT/V3 to simulate the heat transfer in a vertical cylindrical pool of volumetrically heated liquid is estimated. The methodological approach for the estimation implements the ASME V&V 20 standard and includes uncertainty analysis. The study relies on simulation results of four experiments at the BAFOND facility. In the simulation of these experiments, the maximum fluid temperature in a pool is considered as major parameter. The implementation of the approach based on the ASME V&V 20 standard demonstrates that the “traditional” approach for the code accuracy estimation based on calculating the deviation of the calculated value from the measured one with subsequent statistical processing, may result in the underestimation of the modeling error interval.

Soil C-N-P pools and stoichiometry as affected by intensive management of camellia oleifera plantations.

Intensive management of C. oleifera has produced many pure C. oleifera plantations. The transmission of C. oleifera plantation will potentially affect soil C, N, and P pools as well as their stoichiometric characteristics both in top soil layer and vertical soil profile due to the intensive management. To understand changes in vertical pools and stoichiometric characteristics of soil C, N, and P as affected by intensive management of C. oleifera plantations, both mixed and pure C. oleifera plantations were studied. We conducted studies in five locations in Jiangxi, China with both pure and mixed C. oleifera plantations, to compare changes in vertical pools and stoichiometry of C, N, and P. Both C and N pools were significantly different between mixed and pure plantation types of C. oleifera. However, the ratio of C:N, C:P, and N:P was consistently higher in mixed plantations with C:P and N:P altered but C:N ratio did not change with soil depth. The intensive management significantly impact both C and N pools and the stoichiometry of C, N, and P. Intensive management of C. oleifera plantations decreased both C and N pools, especially at the depth of 30–50 cm soil layer. C. oleifera plantation alteration from mixed to pure should be considered in future forest management practice considering the substantial effects on soil element cycling and distribution along vertical soil profile.

MarsNet: Multi-Label Classification Network for Images of Various Sizes

Since the Convolutional Neural Network (CNN) has surfaced and fascinated the world, many researchers have exploited CNN for image classification, object detection, semantic segmentation, etc. However, the conventional CNNs have a pyramidal structure and were designed to process images which have the same size. Although some CNNs can accept images of various sizes, performance is degraded for images smaller than the size of images used for training. In this paper, we propose MarsNet, a CNN based end-to-end network for multi-label classification with an ability to accept various size inputs. In order to allow the network to accept such images, dilated residual network (DRN) is modified to get higher resolution feature maps, and horizontal vertical pooling (HVP) is newly designed to efficiently aggregate positional information from the feature maps. Furthermore, multi-label scoring module and threshold estimation module are employed to serve the purpose of multi-label classification. We verify the effectiveness of the proposed network through two distinctive experiments. We first verify our model by inspecting and classifying multiple types of defects occurred in PCB screen printer using solder paste inspection (SPI) datasets. Secondly, we verify our network using VOC 2007 dataset. Our network is pioneering in that no research has attempted to accomplish multi-label classification for defects in addition to being able to take input images of various sizes in SPI field.

Tracking a Long-Lasting Outer Tropical Cyclone Rainband: Origin and Convective Transformation

AbstractThis study used radar and surface observations to track a long-lasting outer tropical cyclone rainband (TCR) of Typhoon Jangmi (2008) over a considerable period of time (~10 h) from its formative to mature stage. Detailed analyses of these unique observations indicate that the TCR was initiated on the eastern side of the typhoon at a radial distance of ~190 km as it detached from the upwind segment of a stratiform rainband located close to the inner-core boundary. The outer rainband, as it propagated cyclonically outward, underwent a prominent convective transformation from generally stratiform precipitation during the earlier period to highly organized, convective precipitation during its mature stage. The transformation was accompanied by a clear trend of surface kinematics and thermodynamics toward squall-line-like features. The observed intensification of the rainband was not simply related to the spatial variation of the ambient CAPE or potential instability; instead, the dynamical interaction between the prerainband vertical shear and cold pools, with progression toward increasingly optimal conditions over time, provides a reasonable explanation for the temporal alternation of the precipitation intensity. The increasing intensity of cold pools was suggested to play an essential role in the convective transformation for the rainband. The propagation characteristics of the studied TCR were distinctly different from those of wave disturbances frequently documented within the cores of tropical cyclones; however, they were consistent with the theoretically predicted propagation of convectively generated cold pools. The convective transformation, as documented in the present case, is anticipated to be one of the fundamental processes determining the evolving and structural nature of outer TCRs.

Vehicle Re-Identification Using Quadruple Directional Deep Learning Features

In order to resist the adverse effect of viewpoint variations for improving vehicle re-identification performance, we design quadruple directional deep learning networks to extract quadruple directional deep learning features (QD-DLF) of vehicle images. The quadruple directional deep learning networks are with similar overall architecture, including the same basic deep learning architecture but different directional feature pooling layers. Specifically, the same basic deep learning architecture is a shortly and densely connected convolutional neural network to extract basic feature maps of an input square vehicle image in the first stage. Then, the quadruple directional deep learning networks utilize different directional pooling layers, i.e., horizontal average pooling (HAP) layer, vertical average pooling (VAP) layer, diagonal average pooling (DAP) layer and anti-diagonal average pooling (AAP) layer, to compress the basic feature maps into horizontal, vertical, diagonal and anti-diagonal directional feature maps, respectively. Finally, these directional feature maps are spatially normalized and concatenated together as a quadruple directional deep learning feature for vehicle re-identification. Extensive experiments on both VeRi and VehicleID databases show that the proposed QD-DLF approach outperforms multiple state-of-the-art vehicle re-identification methods.

Sonographic Assessment of Amniotic Fluid Volume using Single Deepest Pocket (SDP) among Igbo Women with Uncomplicated Singleton Pregnancies.

Background: The amniotic fluid is fundamental for proper fetal development and growth. Ultrasound visualization of the amniotic fluid permits both subjective and objective estimates of the amniotic fluid. Objective: The objective of this study was to determine the reference values of normal single deepest pocket (SDP) – upper and lower limits, mean SDP and variation of the SDP with gestational age among Igbo women of South-Eastern Nigeria extraction carrying uncomplicated singleton pregnancy. Methodology: This was a prospective cross sectional study involving 400 women carrying uncomplicated singleton pregnancies and who were sure of the date of the first day of their last menstrual period. The single deepest pocket / maximum vertical pool were determined once at presentation at the hospital.. The study was conducted from January 1st to December 31st 2015. The second author carried out all the scanning. The SDP was obtained. Results: The womens’ mean and median ages were the same at 28 years. The gestational age range of the pregnancies was 14-41 weeks. The mean SDP was 5.8cm, while the 5th and 95th percentiles were 3.3cm and 8.5cm respectively. There was no difference in the mean SDP in both term and preterm. There was irregular but continuous rise of mean SDP to a peak of 6.8cm at gestational age of 39 weeks. In conclusion, the participants had a mean SDP of 5.8cm. There was also a positive correlation between SDP and Gestational age. We therefore recommend a longitudinal study to assess perinatal outcome and abnormal amniotic fluid volume among Igbo women of South-Eastern Nigeria. Key words: Single Deepest Pocket, Uncomplicated Singleton Pregnancy, Igbo Women.

Genome sequence analysis of the Indian strain Mannheimia haemolytica serotype A2 from ovine pneumonic pasteurellosis

Mannheimia haemolytica is a leading causative agent of pasteurellosis in ruminants. Genome of M. haemolytica strains from different hosts has been sequenced worldwide to understand its pathogenesis. There are only few reports on the isolation of M. haemolytica in India with limited information on its molecular characteristics. The present study focuses on genome sequence analysis of a M. haemolytica strain isolated from pneumonic sheep. Mannheimia haemolytica A2 strain NIVEDI/MH/1 was isolated and identified by species and serotype-specific PCRs. Whole genome sequencing was performed using the Ion Torrent Personal Genome Machine. A comparative genomic analysis was performed to understand the virulence determinants of the Indian strain and its phylogenetic relationship with other global strains. Sequence data revealed a draft genome of 2,211,426 bp size with 41.3% GC content, assembled into 17 contigs, and contained 2379 genes. Five genomic islands identified in the genome showed high sequence identity with other respiratory pathogens of the Pasteurellaceae family. Phylogenetic analysis showed M. haemolytica A2 NIVEDI/MH/1 is very close to a M. haemolytica A2 strain from pneumonic calf. Further, the analysis revealed the presence of virulence, metal-, and multidrug resistance genes needed for pathogenesis and survival of the bacteria during infection. Also, we identified the presence of type I-C and type II-C of CRISPR-Cas arrays in the present sequenced genome. The study emphasizes the role of M. haemolytica in respiratory infections of ruminants in the Indian subcontinent and indicates the role of vertical and horizontal gene pools in pathogenicity and survivability of the bacteria.

Maternal hemodynamics, fetal biometry and Doppler indices in pregnancies followed up for suspected fetal growth restriction.

To assess whether, in a cohort of patients with a small-for-gestational-age (SGA) fetus with estimated fetal weight ≤ 10th percentile, maternal hemodynamics, fetal biometry and Doppler indices at presentation can predict the subsequent development of an abnormal fetal Doppler index or delivery of a baby with birth weight < 3rd percentile. This was a prospective observational cohort study conducted at a specialist clinic for the management of pregnancies with a SGA fetus at King's College Hospital, London, UK. The study population comprised 86 singleton pregnancies with a SGA fetus, presenting at a median gestational age of 32 (range, 26-35) weeks. We measured maternal cardiac function using a non-invasive transthoracic bioreactance monitor, as well as mean arterial pressure, fetal biometry, and umbilical artery (UA), fetal middle cerebral artery (MCA) and uterine artery (UtA) pulsatility indices (PI), and the deepest vertical pool of amniotic fluid. Z-scores of these variables were calculated based on reported reference ranges and the values were compared between pregnancies with evidence of an abnormal fetal Doppler index at presentation (Group 1), those that had developed an abnormal Doppler index at a subsequent visit (Group 2) and those that did not develop an abnormal Doppler index throughout pregnancy (Group 3). Abnormal fetal Doppler was defined as UA-PI > 95th percentile and/or MCA-PI < 5th percentile. Differences in measured variables at presentation were also compared between pregnancies delivering a baby with birth weight < 3rd percentile and those delivering a baby with birth weight ≥ 3rd percentile. Multivariate logistic regression analysis was used to determine significant predictors of birth weight < 3rd percentile and evolution from normal to abnormal fetal Doppler. In the study population, 14 (16%) cases were in Group 1, 19 (22%) in Group 2 and 53 (62%) in Group3. Birth weight was < 3rd percentile in 39 (45%) cases and ≥ 3rd percentile in 47 (55%). There was decreased cardiac output and stroke volume and increased peripheral vascular resistance compared with a normal population, and the deviations from normal were most marked in Group 1. Pregnancies with birth weight < 3rd percentile, compared with those with birth weight ≥ 3rd percentile, had greater deviations from normal in fetal biometry, maternal cardiac output, stroke volume, heart rate, peripheral vascular resistance and UtA-PI. Multivariate logistic regression analysis demonstrated that, in the prediction of birth weight <3rd percentile, maternal hemodynamic profile provided significant improvement to the prediction provided by maternal demographics, fetal biometry, UtA-PI, UA-PI and MCA-PI (difference between areas under receiver-operating characteristics curves, 0.18 (95% CI, 0.06-0.29); P= 0.002). In contrast, there was no significant independent contribution from maternal hemodynamics in the prediction of the subsequent development of abnormal fetal Doppler. In pregnancies with a SGA fetus, there is decreased maternal cardiac output and stroke volume and increased peripheral vascular resistance, and the deviations from normal are most marked in cases of redistribution in the fetal circulation and reduced amniotic fluid volume. Copyright © 2018 ISUOG. Published by John Wiley & Sons Ltd.

Transfer deep convolutional activation-based features for domain adaptation in sensor networks

In this paper, we propose a novel method named transfer deep convolutional activation-based features (TDCAF) for domain adaptation in sensor networks. Specifically, we first train a siamese network with weight sharing to map the images from different domains into the same feature space, which can learn domain-invariant information. Since various feature maps in one convolutional layer of the siamese network contain different kinds of information, we propose a novel vertical pooling strategy to aggregate them into one convolutional activation summing map (CASM) which contains the completed information and preserves the spatial information. We stretch the CASM into one feature vector to obtain the TDCAF. Finally, we feed the proposed TDCAF into a Support Vector Machine (SVM) for classification. The proposed TDCAF is validated on three generalized image databases and three cloud databases, and the classification results outperform the other state-of-the-art methods.

Comparison of Amniotic Fluid Index and Single Deepest Vertical Pool method for predicting fetal outcome

Background &amp; Objectives: The measurement of amniotic fluid volume (AFV) has been an important component of antenatal evaluation of fetal well being. The most commonly used methods are by amniotic fluid index technique (AFI) and single deepest pool method (SDVP). Both the methods employ ultrasound for AFV measurement. The objective of this study is to compare the usefulness of AFI and SDVP method in assessing amniotic fluid volume for predicting adverse perinatal outcome.Materials &amp; Methods: This is a hospital based prospective comparative study. The patient whose amniotic fluid was measured by SDVP technique was study group and that by AFI was comparison group. Any case with high risk factor was excluded from the study. First case was decided by lottery and then alternately one case was kept in study group and the other in comparative group. If the women did not deliver within 1 week the measurements was retaken and the final values was used for analysis.AFI was estimated as described by Phelan and colleagues and SDVP as described by Manning. Data collection was started after the approval of Institution review board. This study was conducted at Patan Academy of Health Sciences. Duration of data collection was 3 months, Asadh –Bhadra, 2071.Results: A total of one hundred and fifty four cases which met the inclusion criteria were taken. Incidence of oligohydramnios by SDVP method was 10.4 % by SDVP technique and by AFI method it was 18.2 %. There was no statistical significant difference between the two groups in terms of rate of induction, mode of delivery, meconium staining of liquor, fetal heart rate tracings, APGAR score at 5 mins and admission to special care baby unit.Conclusion: In non-high risk pregnancy AFI technique detects slightly more number of oligohydramnios as compared to SDVP technique without apparently any difference in perinatal outcome.

Self‐formed waterfall plunge pools in homogeneous rock

AbstractWaterfalls are ubiquitous, and their upstream propagation can set the pace of landscape evolution, yet no experimental studies have examined waterfall plunge pool erosion in homogeneous rock. We performed laboratory experiments, using synthetic foam as a bedrock simulant, to produce self‐formed waterfall plunge pools via particle impact abrasion. Plunge pool vertical incision exceeded lateral erosion by approximately tenfold until pools deepened to the point that the supplied sediment could not be evacuated and deposition armored the pool bedrock floor. Lateral erosion of plunge pool sidewalls continued after sediment deposition, but primarily at the downstream pool wall, which might lead to undermining of the plunge pool lip, sediment evacuation, and continued vertical pool floor incision in natural streams. Undercutting of the upstream pool wall was absent, and our results suggest that vertical drilling of successive plunge pools is a more efficient waterfall retreat mechanism than the classic model of headwall undercutting and collapse in homogeneous rock.

オオミジンコによる水中物体搬送作業の検討

We have been investigating how to use behavior controlled Daphnia as a living micromachine The Daphnia's control was done by applying their phototaxes. In this study, we developed underwater operation tool for Daphnia. We installed an operational tool called 'Y-Wing custom' to the Daphnia's back and investigated the vertical object transportation performance of the tool installed Daphnia. In the beginning of the development, the attached tool disturbed Daphnia's swimming, and the object transportation performance of the Daphnia with operational tool was not improved comparing to the Daphnia without any tool. After the improvement of the tool shape, we succeeded to move and drop a nylon ball on the shelf from the designated end point of the shelf by manually controlled Daphnia in the vertical plane pool.

The use of amniotic fluid index vs deepest vertical pool in the Maltese population

Introduction: Both AFI and DVP are measures of the amniotic fluid volume antenatally and are an indicator of foetal well-being. The more measurements an index requires, the more risks of error and unnecessary interventions. Aim: To check if there is a difference in using AFI vs DVP. Methods: 46 low risk antenatal patients with singleton pregnancies were evaluated for AFI and DVP between 36 and 40 weeks of gestation. Their pregnancy outcome was noted and results analysed using the unpaired student T-test. Results: Out of the 46 patients, 39 delivered normo-vaginally between 38 and 41weeks with 2 requiring instrumental delivery due to prolonged second stage and 5 requiring an emergency caesarean section 4 for no progress and 1 for foetal distress. All babies were born with good Apgars at 10 min from birth. Comparing AFI to DVP, AFI varied from 5.3 to 24.86 cm with a mean of 13.91 cm (SD: 1.107). DVP varied from 2.54 to 7.94 cm with a mean 4.54 cm (SD: 2.845). P-value was <0.001. Conclusion: There was no difference between AFI and DVP indices making both suitable options to measure amniotic fluid index.

Numerical Modeling for Periodic Oscillation of Free Overfall in a Vertical Drop Pool

AbstractThis study numerically investigates the effect of various plunge pool lengths on the flow conditions with a fixed subcritical approach flow because the flow regimes of free overfall in a vertical drop pool can be mainly classified into skimming, nappe, and periodic oscillatory flows depending upon different discharges and pool dimensions. The simulated water surface and oscillatory period compare well with the experimental data. After the correctness of the numerical model is validated, the velocity and vorticity contours are examined to explore the surface jet of the skimming flow, the impinging jet of the nappe flow, and the dropping flow switching between these two jets. The effect of an air pocket underneath the dropping jet on the oscillatory behavior is addressed, as well as the kinetic and potential energy evolutions are discussed in view of control volume analysis during the periodic oscillatory process. Finally, the authors calculate the hydraulic head loss to evaluate the energy dissipat...

レーザ光を用いたミドリムシ集団の行動制御

In this paper, we investigated to drive a seesaw by automatically position controlled downward flow of the Euglena's bioconvection. The position of the downward flow was controlled by applying Euglena's phototaxis to the scanning blue laser beam. Formally, we succeeded to drive a seesaw by electrically position controlled downward flow of the Tetrahymena's bioconvection. In this paper, we aimed to reproduce the laser control version of this seesaw driving by using Euglena. We have already confirmed that if the experiment is done in the horizontal planar pool, we could make any shapes of Euglena group and we could also assemble simple mechanical parts by pushing the part by Euglena group. And in the vertical pool, controlling position of the bioconvection downward flow was done by irradiating from top surface by scanning blue laser. We succeeded to drive the seesaw by this method. In this study, we investigated the automatic laser controlling algorithm for driving seesaw. At first, we developed algorithm considering the position of the blue laser and the angle of the seesaw. Although driving the seesaw itself was succeeded, driving frequency was less than manual controlling. Therefore, we changed the algorithm that assumes the position of the bioconvective downward flow from the RGB values of the experimental images in the pool for controlling the laser position. By this new algorithm, we succeeded to drive the seesaw faster than the former algorithm and close to the driving time in manual controlling.