•A lightweight tomato target detection algorithm based on YOLOv5 is proposed.•The improved model has a significantly higher detection speed on the CPU platform.•The model is deployed on smartphones ...using model quantization and an APP is developed to achieve local real-time detection of tomatoes.•The 16-bit quantized model performs the best and the detection speed meets the needs of real-time.
The current deep-learning-based tomato target detection algorithm has many parameters; it has drawbacks of large computation, long time consumption, and reliance on high-computing-power devices such as graphics processing units (GPU). In this study, we propose a lightweight improved YOLOv5 (You Only Look Once) based algorithm to achieve real-time localization and ripeness detection of tomato fruits. Initially, this algorithm used a down-sampling convolutional layer instead of the original focus layer, reconstructing the backbone network of YOLOv5 using the bneck module of MobileNetV3. Then, it performs channel pruning for the neck layer to further reduce the model size and uses a genetic algorithm for hyperparameter optimization to improve detection accuracy. We evaluate the improved algorithm using a homemade tomato dataset. The experimental results demonstrated that the improved model number of parameters and floating point operations per second (FLOPs) were compressed by 78% and 84.15% compared to the original YOLOv5s, while the mAP reached 0.969. Meanwhile, the detection speed on the central processing unit (CPU) platform was 42.5 ms, which was 64.88% better. This study further utilized the Nihui convolutional neural network (NCNN) framework to quantize the improved model and developed an Android-based real-time tomato monitoring application (app). Experimental results demonstrated that the 16-bit quantized model achieved an average detection frame rate of 26.5 frames per second (fps) on the mobile side with lower arithmetic power, which was 268% better than the original YOLOv5s, and the model size was reduced by 51.1% while achieving a 93% true detection rate.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
The flavor and market value of loquats are substantially affected by their soluble solids content (SSC). This study explored the ability of hyperspectral imaging combined with multiple regression ...models to detect SSC of loquats using small sample size. In this study, one hundred and fifteen loquats were collected as samples, and regions of interest in the hyperspectral images of the samples were extracted. The spectral data processed through direct orthogonal signal correction (DOSC), the best among the five processing methods examined, was selected for the feature spectrum process. Successive projection algorithm (SPA), competitive adaptive reweighted sampling (CARS), improved uninformative variable elimination algorithm (imUVE), and imUVE-SPA algorithms were used to extract feature spectra and establish partial least squares regression (PLSR), support vector machine regression (SVR), back-propagation neural network (BPNN), and customized convolutional neural network (CNN) models, respectively. The results demonstrated that the CNN model based on the 18 feature spectra extracted by CARS was the best (correlation coefficient of validation = 0.904, root mean square error of validation = 0.336 %, and residual predictive deviation = 2.339). Consequently, hyperspectral images with small sample size combined with appropriate regression models can be used for nondestructive detection of loquat brix. Furthermore, the wavelength feature extraction algorithm effectively improves the performance of CNN models with small sample sizes. This study can serve as a valuable reference for the nondestructive detection of fruit brix in other limited sample situations.
Display omitted
•Hyperspectral imaging can predict the SSC of loquat from small sample sizes.•The performance of customized CNN was the best in SSC prediction.•Feature selection algorithms could improve the performance of the CNN models.•Deep learning can also perform well in the regression problem.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Ultrasound stimulation is increasingly used to investigate brain function and treat brain diseases due to its high level of safety and precise spatiotemporal resolution. Therefore, it is crucial to ...understand the underlying mechanisms involved in ultrasound brain stimulation. In this study, we investigate the role of NMDA receptors in mediating the effects of ultrasound on primary hippocampal neurons in mice. Our results show that ultrasound alone can activate heterologous NMDA receptor subunits, including NR1A, NR2A, and NR2B, in 293T cells, as well as endogenous NMDA receptors in primary neurons. This activation leads to an influx of calcium and an increase in nuclear c-Fos expression in primary neurons that have not been pre-treated with an NMDA receptor inhibitor. In conclusion, our findings demonstrate that NMDA receptors contribute to neuronal activation by ultrasound stimulation in vitro, providing insight into the molecular mechanisms of ultrasound neuromodulation and a new mediator for the sonogenetics technique.
•Low-intensity ultrasound is capable of directly activating exogenously expressed NMDA receptors.•Activation of NMDA receptor contributes Ca2+ influx and c-fos increase induced by ultrasound in primary neurons.•NMDA receptors may be developed as a new mediator for sonogenetics strategy.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Display omitted
•Novel method for antifouling test at high-temperature and high-pressure conditions.•Effects of various factors on antifouling performance of a commercial antifoulant.•Interaction ...mechanisms underlying fouling/antifouling phenomena in SAGD operation.
Oil fouling is commonly encountered in many environmental and engineering fields and remains a challenging issue. The fouling issue in steam-assisted gravity drainage (SAGD) process for extracting bitumen from oil sands deposits has received much attention from oil industry. In view of low capital investment and exceptional antifouling performance, adding antifoulants has emerged as a promising strategy for addressing fouling issues. Developing an efficient method to evaluate their antifouling performance under production conditions is urgently needed. In this work, we report a novel bench-scale test method for quantitatively evaluating the antifouling performance of antifoulants under high temperature and high pressure (HTHP), close to SAGD operation conditions. The effects of antifoulant dosage and stirring rate on the antifouling performance are systematically investigated. The underlying fouling and antifouling mechanisms are characterized through direct molecular and surface force measurements. It is found that the commercial antifoulant exhibits exceptional antifouling performance at the dosage range of 20–35 ppm, and increasing stirring rate enhances the antifouling behavior. The force measurement results indicate that the antifoulant can adsorb to organics and target substrate surfaces, performing the antifouling function by lowering the hydrophobic interaction between organics and target substrates. This work provides a novel and feasible method for characterizing fouling phenomena and evaluating antifoulants under HTHP conditions and improves the understanding of fundamental interaction mechanisms underlying the fouling and antifouling phenomena.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Optogenetics has become a widely used technique in neuroscience research, capable of controlling neuronal activity with high spatiotemporal precision and cell‐type specificity. Expressing exogenous ...opsins in the selected cells can induce neuronal activation upon light irradiation, and the activation depends on the power of incident light. However, high optical power can also lead to off‐target neuronal activation or even cell damage. Limiting the incident power, but enhancing power distribution to the targeted neurons, can improve optogenetic efficiency and reduce off‐target effects. Here, the use of optical lenses made of polystyrene microspheres is demonstrated to achieve effective focusing of the incident light of relatively low power to neighboring neurons via photonic jets. The presence of microspheres significantly localizes and enhances the power density to the target neurons both in vitro and ex vivo, resulting in increased inward current and evoked action potentials. In vivo results show optogenetic stimulation with microspheres that can evoke significantly more motor behavior and neuronal activation at lowered power density. In all, a proof‐of‐concept of a strategy is demonstrated to increase the efficacy of optogenetic neuromodulation using pulses of reduced optical power.
Activation of ChR2 by blue light can be enhanced by polystyrene (PS) microsphere in vitro. PS microsphere facilitates the regulation of neuron activity by opening of ChR2 ex vivo. Photonic nanojet‐mediated optogenetics by PS microsphere improves the regulation of targeting neuronal activity in vivo.
Full text
Available for:
FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
Bitumen froth, as the crude product from the surface mining of oil sands, requires the efficient removal of water and solids for further processing. This work has systematically characterized the ...properties of bitumen froth from an oil sands operation in Northern Alberta and their correlation to the froth treatment efficiency (i.e., removal of water and demulsification behaviors). Systematic characterization was performed regarding the composition, wettability, and particle size distribution of the solids separated from different phases (i.e., diluted bitumen, aqueous phase, and coarse solids). Multiscale analyses including X-ray photoelectron spectroscopy (XPS), scanning electron microscopy coupled with energy dispersive X-ray spectroscopy (SEM/EDX), X-ray powder diffraction (XRD), and elemental analysis were employed to analyze the surface and bulk chemistry of the solids. The results indicate that a great amount of toluene-soluble and toluene-insoluble organics are discontinuously adsorbed on the solid surfaces, making the solid biwettable with a water contact angle of around 90°. The biwettable and fine solids are apt to remain at the oil/water interface or are trapped between the oil droplets, strengthening the “protective” interfacial film, which contributes to the formation of rag layer. The fine and ultrafine solids with a surface coating of organics remain with bitumen and adversely influence the quality of the separated bitumen. The organic-coated coarse solids precipitate toward the bottom, reducing the overall bitumen recovery. In addition, a trace amount of the organics could be firmly adsorbed on the solid surface, which further impedes the demulsification efficiency. This work correlates the characteristics of the bitumen froth and its demulsification efficiency using two commercial demulsifiers, with useful implications for developing effective and cost-efficient demulsifiers and emulsion treatment technologies in oil production and other related chemical and engineering processes.
Full text
Available for:
IJS, KILJ, NUK, PNG, UL, UM
PDF
