Lightweight Fish Bait Particle Counting Method Based on Pruning and Shift Quantization

In the process of aquaculture,monitoring the number of fish bait particles is of great significance to improve the growth and welfare of fish.Although the counting method based on onvolutional neural network(CNN)achieve good accuracy and applicability,it has a high amount of parameters and computation,which limit the deployment on resource-constrained hardware devices.In order to solve the above problems,this paper proposes a lightweight bait particle counting method based on shift quantization and model pruning strategies.Firstly,we take corresponding lightweight strategies for different layers to flexibly balance the counting accuracy and performance of the model.In order to deeply lighten the counting model,the redundant and less informative weights of the model are removed through the combination of model quantization and pruning.The experimental results show that the compression rate is nearly 9 times.Finally,the quantization candidate value is refined by introducing a power-of-two addition term,which improves the matches of the weight distribution.By analyzing the experimental results,the counting loss at 3 bit is reduced by 35.31%.In summary,the lightweight bait particle counting model proposed in this paper achieves lossless counting accuracy and reduces the storage and computational overhead required for running convolutional neural networks.

One-step detection of T4 polynucleotide kinase activity based on single particle-confined enzyme reaction and digital particle counting

T4 polynucleotide kinase(T4 PNK) is a pivotal enzyme for DNA replication, recombination, and DNA damage repair. Herein, a robust single particle counting-based assay has been developed for the high-sensitive determination of T4 PNK activity through only a simple one-step reaction. Taking benefit of the exceptional space-confined enzymatic property of T4 PNK towards DNA substrates on a single nanoparticle,the T4 PNK activity can be precisely determined by counting the fluorescence-positive nanoparticles in a digital manner with a total internal reflection fluorescent microscope(TIRFM). Due to the featured spatial-confined enzymatic property of T4 PNK and the single particle counting-based signal readout, T4PNK can be effectively differentiated from other interfering enzymes. This facile strategy has been also successfully applied to screen T4 PNK inhibitor and accurately determine T4 PNK activity in complex biological samples, paving a potential avenue for the digital analysis of biomarkers.

Effect of a biological activated carbon filter on particle counts

Due to the importance of biological safety in drinking water quality and the disadvantages which exist in traditional methods of detecting typical microorganisms such as Cryptosporidium and Giardia, it is necessary to develop an alternative. Particle counts is a qualitative measurement of the amount of dissolved solids in water. The removal rate of particle counts was previously used as an indicator of the effectiveness of a biological activated carbon (BAC) filter in removing Cryptosporidium and Giardia. The particle counts in a BAC filter effluent over one operational period and the effects of BAC filter construction and operational parameters were investigated with a 10 m3/h pilot plant. The results indicated that the maximum particle count in backwash remnant water was as high as 1296 count/ml and it needed about 1.5 h to reduce from the maximum to less than 50 count/ml. During the standard filtration period, particle counts stay constant at less than 50 count/ml for 5 d except when influ- enced by sand filter backwash remnant water. The removal rates of particle counts in the BAC filter are related to characteristics of the carbon. For example, a columned carbon and a sand bed removed 33.3% and 8.5% of particles, respectively, while the particle counts in effluent from a cracked BAC filter was higher than that of the influent. There is no significant difference among particle removal rates with different filtration rates. High post-ozone dosage (>2 mg/L) plays an important role in particle count removal; when the dosage was 3 mg/L, the removal rates by carbon layers and sand beds decreased by 17.5% and increased by 9.5%, respectively, compared with a 2 mg/L dosage.

High Speed Image Acquisition ＆ Neutral Particle Count Systems and the RTX Research

In HL-2A tokamak experiment, high speed image acquisitoin system grabs images and shows them in real-time. It provides serial images of discharge to help scientists a lot to analyse the experiment. Neutral particle count system counts the number of neutral particles radiated from fusion plasmas. It provides the data to calculate the ion energy.

CFD modeling and laboratory studies of dust cleaning efficacy of an efficient four stage non-clogging impingement filter for flooded-bed dust scrubbers

Fibrous-type flters are used to capture dust particles in mining and other occupations where personnel are exposed for prolonged periods.Dust cleansing devices including fooded-bed dust scrubbers use these mesh-type multi-layered flters.These flters trap dust particles efciently on their surface and inside their mesh.However,their continued operation leads to dust build-up and clogging.This results in increased resistance of the flter and lowered airfow rate through the scrubber.This could potentially enhance the exposure of the miners.A non-clogging self-cleaning impingement screen type dust flter was designed by the authors for use in mining and industrial dust cleansing applications.The flter guides dirt-laden air through rapidly turning paths which forces it to shed heavier particles.The particles impact one of the impermeable solid metallic flter surfaces and are removed from the airstream.A full cone water spray installed upstream prevents any surface buildup of dust.This paper summaried the computer models generated to show the flter operations and laboratory experiments including optical particle counting to establish the cleaning efciency.

Using chloramine as a coagulant aid in enhancing coagulation of Yellow River water in China

Considering that contaminated raw water mostly contains high Ammonia-N and a majority of water treatment plants use prechlorination process in China, efficiency of chloramines as a coagulant aid in enhancing coagulation was investigated by Jar stirring and pilot-scale tests, using Yellow River water containing high concentration of natural organic matters （NOM） and bromide in winter. The jar tests results showed that, compared with no preoxidation, preformed chloramine apparently decreased the turbidity of settled and filtered water with low dosage （2.0 rag/L）, and the aid-coagulation efficiency was further enhanced with the increase of chlorine （Cl2） to Ammonia-N （N） ratio. Pilot-scale studies indicated that, in comparison to the case without preoxidation, the turbidity removal efficiency of flotation and filtration effluent water was significantly improved, the particle counts of filtered water were decreased 63.4%, the average rate of filter head loss was reduced 18.2%, and filter run time was prolonged 15.7%. Therefore, chloramine preoxidation may substantially enhance the particle separation efficiency.