YOLSO: You Only Look Small Object

Small object detection is challenging and far from satisfactory. Most general object detectors suffer from two critical issues with small objects: (1) Feature extractor based on classification network cannot express the characteristics of small objects reasonably due to insufficient appearance information of targets and a large amount of background interference around them. (2) The detector requires a much higher location accuracy for small objects than for general objects. This paper proposes an effective and efficient small object detector YOLSO to address the above problems. For feature representation, we analyze the drawbacks in previous backbones and present a Half-Space Shortcut(HSSC) module to build a background-aware backbone. Furthermore, a coarse-to-fine Feature Pyramid Enhancement(FPE) module is introduced for layer-wise aggregation at a granular level to enhance the semantic discriminability. For loss function, we propose an exponential L1 loss to promote the convergence of regression, and a focal IOU loss to focus on prime samples with high classification confidence and high IOU. Both of them significantly improves the location accuracy of small objects. The proposed YOLSO sets state-of-the-art results on two typical small object datasets, MOCOD and VeDAI, at a speed of over 200 FPS. In the meantime, it also outperforms the baseline YOLOv3 by a wide margin on the common COCO dataset.     

Reduction of waste biosolids by RAS-ozonation: Model validation and sensitivity analysis for biosolids reduction and nitrification

Biosolids reduction model by return activated sludge ozonation was validated by simulating nitrification data compiled from our pilot-scale and the literature studies. Then, a global sensitivity analysis (GSA) was performed to identify influential and non-influential parameters for biosolids reduction efficiency, change in specific nitrification activity (SNA), and alteration to expected nitrification stability. In general, the model outputs were sensitive to operational and ozone reaction parameters, but not to biochemical parameters. For operational parameters, mainly temperature and initial solids retention time (SRT) influenced all model outputs. For biosolids reduction, increase in the degradability of the influent COD decreased the reduction efficiency. For SNA, the changes were highly dependent on the influent TKN/COD ratio. Our findings also imply that the stability of the nitrification process in ozonated systems should be enhanced at constant MLVSS for warm temperatures, but could be reduced at temperatures below 12 °C and aerated SRTs below 10 days.     

Biological Water Treatment by a Hybrid Fluidized-Bed Bioreactor: Theoretical Study

A hybrid fluidized-bed bioreactor for water purification was proposed and analyzed. It is a novel type of bioreactor characterized by hitherto unknown stationary and dynamic features. Steady-state characteristics of this hybrid bioreactor with external liquid circulation are presented. A quantitative analysis of steady-state properties of the bioreactor was performed with the aid of an original mathematical model developed for a double-substrate aerobic microbiological process. A steady-state analysis of aerobic processes characterized by different oxygen demand was performed. The effect of essential parameters was evaluated, including carbonaceous substrate concentration in the feed stream to the apparatus, aeration intensity, total residence time of a liquid in the bioreactor, and height of the bed of fine carrier particles.     

Pilot Plant Study of Simultaneous Sewage Sludge Digestion and Metal Leaching

This research is related to a preindustrial pilot scale study of the performance of the simultaneous sewage sludge digestion and metal leaching (SSDML) process for decontamination and stabilization of sewage sludge. Ten batch tests were carried out in two 4?m3 bioreactors under various conditions of operation. Results indicated that the addition of 1.0 to 1.5?g?S0/L, which is the equivalent of approximately 30 to 50?kg?S0 per tons dry sludge, is sufficient to obtain conditions of acidity (pH<2.5) and oxido-reduction potential greater than 500 mV necessary for an effective solubilization of toxic metals. The final average of metal solubilization in the output sludge during the SSDML process varied in the following ranges: 25–78% Cd, 9–32% Cr, 48–100% Cu, 77–99% Mn, 15–53% Ni, 12–47% Pb, and 66–100% Zn. The N, P, and K contents were also preserved in the decontaminated sludge. Moreover, the use of low concentrations of elemental sulfur makes it possible to obtain decontaminated sludge with a low total sulfur content (1.4–1.5% S) compatible with agricultural use. The suspended solids removal calculated for the SSDML process was slightly lower, (2.5±0.4)% volatile suspended solids per day, than those reached using standard aerobic digestion. Finally, the SSDML process was found to be effective in removing bad odors and in the destruction (99–100%) of indicator bacteria.     

Non-equilibrium particle motion in the vicinity of a single blade

The motion of sand particles close to a single moving blade was investigated using Positron Emission Particle Tracking (PEPT) during the period in which the free bed surface profile was evolving to an equilibrium shape. The area affected by the blade was divided into active and inactive regions and these were analysed separately. The characteristic heart-shape of the active region in the plan view was determined. An approximately 10-particle-diameter wide velocity transition zone is found between the two regions. While the tracer particle is in the inactive region moving away from the blade, the time dependence of the axial displacement is well described by a logarithmic relationship. The probability of particle movement towards the centre of the blade was quantified using a “central tendency” index. The calculated central tendency shows maxima at each side of the blade. The separation of the two maxima, which indicates the width of the active region, increases with fill level but is independent of rotational speed.     

Effect of electric field on the hydrodynamics of fluidized nanoparticles

The effect of electric field on the hydrodynamics of nanoparticles was studied in a fluidized rectangular bed, with electrodes attached to two parallel walls. It was shown that the electric field of the order of 3 times the gravity markedly decreased the bed expansion and increased the solids volume fraction of nanoparticles fluidized by air. In these experiments, a light diode assembly was utilized to infer the local solids volume fractions within a rectangular bed of 10 nm silica particles. These experimental measurements yielded a two dimensional solids volume fraction distribution within the rectangular bed. The experimental results provided some new insights into the distribution of solids within the bed. The agglomerate diameters were computed using a momentum balance with the drag given by the Ergun equation and the empirical Richardson-Zaki method. Both methods yielded agglomerate diameters of the order of 100 μm and showed dependence on the strength of the electric field. The electric field decreased the granular temperature of the nanoparticles.     

Off-Site Exposure to Respirable Aerosols Produced during the Disk-Incorporation of Class B Biosolids

Field experiments were conducted at a Class B biosolids land application site in central Arizona to measure, model, and source-track the off-site transport of aerosols emitted when biosolids were disk-incorporated into soils. Real-time PM10 monitoring provided time-resolved aerosol information sufficient for verifying both off-site concentration and off-site exposure time model results. Under the conditions considered and at a distance of 165?m from the aerosol source, biosolids disk-incorporation resulted in an intermittent exposure to biosolids-derived aerosol concentration between 15 and 40?μg/m3 and an inhalable biosolids dose between 2 and 8?μg. Transport modeling predicted that these doses will decrease with increasing wind speed. In addition, three DNA sequence-based biosolids source tracking methods were applied to aerosol samples and confirmed the presence of biosolids in aerosols at 5, 65, and 165?m from the aerosol source. Field measurements and modeling indicate that the nature of biosolids-derived aerosol exposure is a series of intermittent high concentration puffs, rather than a continuous low concentration.     

Two-Dimensional Physical and Numerical Modeling of a Pile-Supported Earth Platform over Soft Soil

This paper focuses on the mechanisms occurring in a granular earth platform over soft ground improved by rigid piles. Two-dimensional physical model experiments were performed using the Schneebeli’s analogical soil to investigate the load transfer mechanisms by arching and the settlement reduction and homogenization. Experimental outputs are compared to results obtained on a numerical model using a plane strain continuum approach. The impact of the constitutive model complexity to simulate the platform material behavior was first assessed, but no large difference was recorded. As far as the proposed model, which takes the main features of the observed behavior satisfactorily into account, the numerical procedure could be validated and the parametric studies extended numerically. Both approaches of this study underlined the main geometrical and geotechnical parameters which should inevitably be taken into account in a simplified design method, namely the capping ratio, the platform height, and the platform material shear strength.     

The Application of Ozone for Water Treatment In The United Kingdom — Current Practive and Recent Research

Only two water works in the UK apply ozone at present as part of their treatment, onefor taste and odor control and theother forcolor removal. Forthetwo applicationus ozone was the most economical option at thetime theplants were installed andozone has been successful for the purpose for which it was designed.