Search

Crop monitoring focuses on detecting and identifying numerous crops within a limited region. A major challenge arises from the fact that the target crops are typically smaller in size compared to the image resolution, as seen in the case of rice plants. For instance, a rice plant may only span a few dozen pixels in an aerial image that comprises thousands to millions of pixels. This size discrepancy hinders the performance of standard detection methods. To overcome this challenge, our proposed solution includes a testing time grid cropping method to reduce the scale gap between rice plants and aerial images, a multi-scale prediction method for improved detection using cropped images based on varying scales, and a mean-NMS to prevent the potential exclusion of promising detected objects during the NMS stage. Furthermore, we introduce an efficient object detector, the Enhanced CSL-YOLO, to expedite the detection process. In a comparative analysis with two advanced models based on the public test set of the AI CUP 2021, our method demonstrated superior performance, achieving notable 4.6% and 2.2% increases in F1 score, showcasing impressive results. [ABSTRACT FROM AUTHOR]

Path planning is a key problem in the design of autonomous driving systems, and accurate traffic light detection is very important for robust routing. In this paper, we devise an object detection model, which mainly focuses on traffic light classification at a distance. In the past, most techniques employed in this field were dominated by high-intensity convolutional neural networks (CNN), and many advances have been achieved. However, the size of traffic lights may be small, and how to detect them accurately still deserves further study. In the object detection domain, the scheme of feature fusion and transformer-based methods have obtained good performance, showing their excellent feature extraction capability. Given this, we propose an object detection model combining both the pyramidal feature fusion and self-attention mechanism. Specifically, we use the backbone of the mainstream one-stage object detection model consisting of a parallel residual bi-fusion (PRB) feature pyramid network and attention modules, coupling with architectural tuning and optimizer selection. Our network architecture and module design aim to effectively derive useful features aimed at detecting small objects. Experimental results reveal that the proposed method exhibits noticeable improvement in many performance indicators: precision, recall, F1 score, and mAP, compared to the vanilla models. In consequence, the proposed method obtains good results in traffic light detection. [ABSTRACT FROM AUTHOR]

This letter proposes a novel method to detect carried objects from videos and applies it for analysis of suspicious events. First of all, we propose a novel kernel-based tracking method for tracking each foreground object and further obtaining its trajectory. With the trajectory, a novel ratio histogram is then proposed for analyzing the interactions between the carried object and its owner. After color re-projection, different carried objects can be then accurately segmented from the background by taking advantages of Gaussian mixture models. After bag detection, an event analyzer is then designed to analyze various suspicious events from the videos. Even though there is no prior knowledge about the bag (such as shape or color), our proposed method still performs well to detect these suspicious events. As we know, due to the uncertainties of the shape and color of the bag, there is no automatic system that can analyze various suspicious events involving bags (such as robbery) without using any manual effort. However, by taking advantages of our proposed ratio histogram, different carried bags can be well segmented from videos and applied for event analysis. Experimental results have proved that the proposed method is robust, accurate, and powerful in carried object detection and suspicious event analysis. Index Terms--Carried bag detection, finite state machines, gaussian mixture models, ratio histogram, suspicious event detection.

We describe the synthesis and characterization of wurtzite (Ga1−xZnx)(N1−xOx) nanocrystals with a wide range of compositions and a focus on properties relevant for solar fuel generation. (Ga1−xZnx)(N1−xOx), a solid solution of GaN and ZnO, is an intriguing material because it exhibits composition-dependent visible absorption even though the parent semiconductors absorb in the UV. When functionalized with co-catalysts, (Ga1−xZnx)(N1−xOx) is also capable of water splitting under visible irradiation. Here, we examine the synthesis of (Ga1−xZnx)(N1−xOx) nanocrystals to understand how they form by nitridation of ZnO and ZnGa2O4 nanocrystalline precursors. We find that the ZnO precursor is critical for the formation of crystalline (Ga1−xZnx)(N1−xOx) at 650 °C, consistent with a mechanism in which wurtzite (Ga1−xZnx)(N1−xOx) nucleates topotactically on wurtzite ZnO at an interface with ZnGa2O4. Using this information, we expand the range of compositions from previously reported 0.30 ≤x≤ 0.87 to include the low-x and high-x ends of the range. The resulting compositions, 0.06 ≤x≤ 0.98, constitute the widest range of (Ga1−xZnx)(N1−xOx) compositions obtained by one synthetic method. We then examine how the band gap depends on sample composition and find a minimum of 2.25 eV at x = 0.87, corresponding to a maximum possible solar-to-H2 power conversion efficiency of 12%. Finally, we examine the photoelectrochemical (PEC) oxidation behavior of thick films of (Ga1−xZnx)(N1−xOx) nanocrystals with x = 0.40, 0.52, and 0.87 under visible illumination. (Ga1−xZnx)(N1−xOx) nanocrystals with x = 0.40 exhibit solar PEC oxidation activity that, while too low for practical applications, is higher than that of bulk (Ga1−xZnx)(N1−xOx) of the same composition. The highest photocurrents are observed at x = 0.52, even though x = 0.87 absorbs more visible light, illustrating that the observed photocurrents are a result of an interplay of multiple parameters which remain to be elucidated. This set of characterizations provides information useful for future studies of composition-dependent PEC properties of nanoscale (Ga1−xZnx)(N1−xOx). [ABSTRACT FROM AUTHOR]

Using femtosecond transient absorption spectroscopy, we investigated hot carrier distributions in semiconductor cadmium selenide quantum dots. The relaxation processes represent the behavior of an ensemble of QDs. This concept is applied for analysis with the Fermi-Dirac distribution and relaxation processes among different electron-hole pair states. By extracting the experimental hot carrier distribution and fitting with the Fermi-Dirac function, we resolved the rapid thermalization processes, such as carrier-carrier and carrier-phonon interactions was resolved within one picosecond upon photoexcitation. The analysis, using the Fermi-Dirac distribution modulated by the density of states, provides a general route to understanding the carrier cooling and heat dissipation processes in quantum dot-based systems. [ABSTRACT FROM AUTHOR]

For over half a century, alternating electric fields have been used to induce particle transport, furnishing the ζ-potential of analytes with sizes ranging from a few nanometers to several micrometers. Concurrent advances in nanotechnology have provided new materials for catalysis, self-assembly, and biomedical applications, all of which benefit from a thorough understanding of particle surface charge. [ABSTRACT FROM AUTHOR]

Mangroves are grown in intertidal zones along tropical and subtropical climate areas, which have many benefits for humans and ecosystems. The knowledge of mangrove conditions is essential to know the statuses of mangroves. Recently, satellite imagery has been widely used to generate mangrove and degradation mapping. Sentinel-2 is a volume of free satellite image data that has a temporal resolution of 5 days. When Hurricane Irma hit the southwest Florida coastal zone in 2017, it caused mangrove degradation. The relationship of satellite images between pre and post-hurricane events can provide a deeper understanding of the degraded mangrove areas that were affected by Hurricane Irma. This study proposed an MDPrePost-Net that considers images before and after hurricanes to classify non-mangrove, intact/healthy mangroves, and degraded mangroves classes affected by Hurricane Irma in southwest Florida using Sentinel-2 data. MDPrePost-Net is an end-to-end fully convolutional network (FCN) that consists of two main sub-models. The first sub-model is a pre-post deep feature extractor used to extract the spatial–spectral–temporal relationship between the pre, post, and mangrove conditions after the hurricane from the satellite images and the second sub-model is an FCN classifier as the classification part from extracted spatial–spectral–temporal deep features. Experimental results show that the accuracy and Intersection over Union (IoU) score by the proposed MDPrePost-Net for degraded mangrove are 98.25% and 96.82%, respectively. Based on the experimental results, MDPrePost-Net outperforms the state-of-the-art FCN models (e.g., U-Net, LinkNet, FPN, and FC-DenseNet) in terms of accuracy metrics. In addition, this study found that 26.64% (41,008.66 Ha) of the mangrove area was degraded due to Hurricane Irma along the southwest Florida coastal zone and the other 73.36% (112,924.70 Ha) mangrove area remained intact. [ABSTRACT FROM AUTHOR]

Radar automatic target recognition is a critical research topic in radar signal processing. Radar high-resolution range profiles (HRRPs) describe the radar characteristics of a target, that is, the characteristics of the target that is reflected by the microwave emitted by the radar are implicit in it. In conventional radar HRRP target recognition methods, prior knowledge of the radar is necessary for target recognition. The application of deep-learning methods in HRRPs began in recent years, and most of them are convolutional neural network (CNN) and its variants, and recurrent neural network (RNN) and the combination of RNN and CNN are relatively rarely used. The continuous pulses emitted by the radar hit the ship target, and the received HRRPs of the reflected wave seem to provide the geometric characteristics of the ship target structure. When the radar pulses are transmitted to the ship, different positions on the ship have different structures, so each range cell of the echo reflected in the HRRP will be different, and adjacent structures should also have continuous relational characteristics. This inspired the authors to propose a model to concatenate the features extracted by the two-channel CNN with bidirectional long short-term memory (BiLSTM). Various filters are used in two-channel CNN to extract deep features and fed into the following BiLSTM. The BiLSTM model can effectively capture long-distance dependence, because BiLSTM can be trained to retain critical information and achieve two-way timing dependence. Therefore, the two-way spatial relationship between adjacent range cells can be used to obtain excellent recognition performance. The experimental results revealed that the proposed method is robust and effective for ship recognition. [ABSTRACT FROM AUTHOR]

Recently, digital learning has attracted a lot of researchers to improve the problems of learning carelessness, low learning ability, lack of concentration, and difficulties in comprehending the logic of math. In this study, a digital learning system based on Kinect somatosensory system is proposed to make children and teenagers happily learn in the course of the games and improve the learning performance. We propose two interactive geometry and puzzle games. The proposed somatosensory games can make learners feel curious and raise their motivation to find solutions for boring problems via abundant physical expressions and interactive operations. The players are asked to select particular operation by gestures and physical expressions within a certain time. By doing so, the learners can feel the fun of game playing and train their logic ability before they are aware. Experimental results demonstrate that the proposed somatosensory system can effectively improve the students' learning performance. [ABSTRACT FROM AUTHOR]

Results of our study on ultrafast electron transfer (eT) dynamics from coumarins (coumarin-1, coumarin-480, and coumarin-153) incarcerated within octa acid (OA) capsules as electron donors to methyl viologen dissolved in water as acceptor are presented. Upon photoexcitation, coumarin inside the OA capsule transfers an electron to the acceptor electrostatically attached to the capsule leading to a long-lived radical-ion pair separated by the OA capsular wall. This charge-separated state returns to the neutral ground state via back electron transfer on the nanosecond time scale. This system allows for ultrafast electron transfer processes through a molecular wall from the apolar capsular interior to the highly polar (aqueous) environment on the femtosecond time scale. Employing femtosecond transient absorption spectroscopy, distinct rates of both forward (1-25 ps) and backward eT (700-1200 ps) processes were measured. Further understanding of the energetics is provided using Rehm-Weller analysis for the investigated photoinduced eT reactions. The results provide the rates of the eT across a molecular wall, akin to an isotropic solution, depending on the standard free energy of the reaction. The insights from this work could be utilized in the future design of efficient electron transfer processes across interfaces separating apolar and polar environments.

Developing type-II heterostructures with a spatial separation of photoexcited electrons and holes is a useful route to promote photocatalytic hydrogen generation. However, few investigations on the charge transfer process across the heterojunction have been carried out, which can allow us to uncover the reaction mechanism. Herein, CdSe quantum dots (QDs) and TiO 2 nanocrystals were synthesized and combined in water yielding CdSe/TiO 2 type II heterostructures. It was found that mercaptopropionic acid as bifunctional molecules could bind with CdSe and TiO 2 to form a cross-linked morphology. The charge carrier dynamics of bare CdSe and CdSe/TiO 2 were detected using femtosecond transient absorption spectroscopy. In the presence of TiO 2 , the average exciton lifetime of CdSe QDs was apparently decreased, owing to the electron transfer from photoexcited CdSe to TiO 2 . Particularly, the electron-transfer rate from small CdSe QDs (3.0 nm) was much faster than that from big CdSe QDs (4.2 nm). The improved photocatalytic hydrogen generation was observed for CdSe/TiO 2 compared to bare CdSe QDs. The enhancement factor for small CdSe QDs was higher than that for big CdSe QDs, which was in good agreement with the electron-transfer rates. This result indicated that the electron transfer between CdSe and TiO 2 played an important role in photocatalytic hydrogen generation on CdSe/TiO 2 type-II heterostructure. Our study provides a fundamental guidance to construct efficient heterostructured photocatalysts by delicate control of the band alignment.

(Ga(1-x)Zn(x))(N(1-x)O(x)) is a visible absorber of interest for solar fuel generation. We present a first report of soluble (Ga(1-x)Zn(x))(N(1-x)O(x)) nanocrystals (NCs) and their excited-state dynamics over the time window of 10(-13)-10(-4) s. Using transient absorption spectroscopy, we find that excited-state decay in (Ga0.27Zn0.73)(N0.27O0.73) NCs has both a short (<100 ps) and a long-lived component, with a long overall average lifetime of ∼30 μs. We also find that the strength of the visible absorption is comparable to that of direct band gap semiconductors such as GaAs. We discuss how these results may relate to the origin of visible absorption in (Ga(1-x)Zn(x))(N(1-x)O(x)) and its use in solar fuel generation.

Recently, digital learning has attracted a lot of researchers to improve the problems of learning carelessness, low learning ability, lack of concentration, and difficulties in comprehending the logic of math. In this study, a digital learning system based on Kinect somatosensory system is proposed to make children and teenagers happily learn in the course of the games and improve the learning performance. We propose two interactive geometry and puzzle games. The proposed somatosensory games can make learners feel curious and raise their motivation to find solutions for boring problems via abundant physical expressions and interactive operations. The players are asked to select particular operation by gestures and physical expressions within a certain time. By doing so, the learners can feel the fun of game playing and train their logic ability before they are aware. Experimental results demonstrate that the proposed somatosensory system can effectively improve the students' learning performance.

Supramolecular photoinduced electron transfer dynamics between coumarin 153 (C153) and 4,4'-dimethyl viologen dichloride (MV(2+)) across the molecular barrier of a host molecule, octa acid (OA), has been investigated with femtosecond time resolution. The ultrafast electron transfer from C153 to MV(2+) followed excitation with 150 fs laser pulses at a wavelength of 390 nm despite the fact that C153 was incarcerated within an OA(2) capsule. As a result, the photoexcited coumarin did not show any of the typical relaxation dynamics that is usually observed in free solution. Instead, the excited electron was transferred across the molecular wall of the capsuleplex within 20 ps. Likewise, the lifetime of the charge transfer state was short (724 ps), and electron back-transfer reestablished the ground state of the system within 1 ns, showing strong electronic coupling among the excited electron donor, host, and acceptor. When the donor was encapsulated into the host molecule, the electron transfer process showed significantly accelerated dynamics and essentially no solvent relaxation compared with that in free solution. The study was also extended to N-methylpyridinium iodide as the acceptor with similar results.

Femtosecond laser spectroscopy has now been a powerful technique for over a decade to investigate charge carrier dynamics in nanoscale optoelectronic systems with a temporal resolution of 100 fs (10(-13) s) or better. Both transient absorption and time-resolved photoluminescence spectroscopy are now popular spectroscopic techniques, which are well-established and provide direct insight into the charge carrier dynamics of nanomaterials. In this Perspective, we focus mainly on the developments with regard to studies of semiconductor nanostructures. Controlling the charge carrier dynamics, including hot carrier relaxation, trapping, interfacial carrier transfer, carrier multiplication, and recombination, is essential for successful energy conversion or photocatalysis, to name two major optoelectronic applications. We will show how femtosecond laser spectroscopy evolved into techniques that unveil the dynamic charge carrier properties of semiconductor nanomaterials toward heterostructures and complex nanoarchitectures and that femtosecond time-resolved laser spectroscopy can shine light on the path to novel optoelectronic structures and emergent optoelectronic technologies.

Using femtosecond transient absorption and time-resolved photoluminescence spectroscopy, we studied the electron versus hole dynamics in photoexcited quasi-type-II heterostructured nanocrystals with fixed CdTe core radii and varying CdSe shell coverage. By choosing the pump wavelength in resonance with the core or the shell states, respectively, we were able to measure the excited electron and hole dynamics selectively. Both, the core- and the shell-excited CdTe/CdSe nanocrystals showed the same spectral emission and photoluminescence lifetimes, indicating that ultrafast electron and hole transfer across the core/shell interface resulted in the identical long-lived charge transfer state. Both charge carriers have subpicosecond transfer rates through the interface, but the subsequent relaxation rates of the hole (τ(dec) ∼ 800 ps) and electron (τ(avg) ∼ 8 ps) are extremely different. On the basis of the presented transient absorption measurements and fitting of the steady-state spectra, we find that the electron transfer occurs in the Marcus inverted region and mixing between the CdTe exciton and charge transfer states takes place and therefore needs to be considered in the analysis.