During the growth stage of soybean seedlings, it is crucial to quickly and precisely identify them for emergence rate assessment field management. Traditional manual counting methods have some limitations in scenarios with large-scale high-efficiency requirements, such as being time-consuming, labor-intensive, prone human error (such subjective judgment visual fatigue). To address these issues, this study proposes a rapid detection method suitable airborne edge devices dense seedling images....

The foreign object debris (FOD) on the airport runways is a serious threat to aircraft taxiing, taking off and landing. installation of cameras side track automatically monitor FOD very important for improving safety runway. This paper proposes detection method based Convolutional Neural Network (CNN) small-scale FOD. employs FasterR-CNN framework generate candidate regions input image. uses DenseNet instead traditional VGGl6Net feature extraction, which can greatly reduce network parameters...

Weather recognition is of great significance in traffic safety, environment and meteorology. However, the visual image features weather are highly abstractive, traditional method has a high computational complexity low accuracy. In this paper, edge deterioration phenomenon introduced convolution neural network (CNN) to solve problem that common CNN cannot distinguish specific weather. The proposed used Mask R-CNN extract regions interest including foreground edges image, superimposed them...

Abstract Low visibility always leads to serious traffic accidents worldwide, and it remains a tough problem despite the extensive work researched in field of meteorology. Therefore, we propose novel end-to-end network, FGS-Net, for estimation, combining "engineered features" "learned achieve higher accuracy. Specifically, effective fog region segmentation method, named ASRS(Auto Seed Region Segmentation), segment regions input image. Subsequently, two "specific (transmittance matrix, dark...

Low visibility always leads to serious traffic accidents worldwide, although extensive works are studied deal with the estimation of in meteorology areas, it is still a tough problem.Deep learning-based methods, has low accuracy due lacking "specific features" foggy images.Meanwhile, physical modelbased methods only applicable some specific scenes its high requirements for extra auxiliary parameters.Therefore, This paper proposes novel end-to-end framework named STCN-Net estimation, which...

Abstract An efficient and gram‐scale enantioselective synthesis of ( R )‐ S )‐13‐methylheptacosane, the sex pheromone pear psylla, has been developed. The key steps approach included Evans' chiral auxiliaries Wittig coupling phosphonium salt with aldehyde.

Compared with expensive pixel-wise annotations, image-level labels make it possible to learn semantic segmentation in a weakly-supervised manner. Within this pipeline, the class activation map (CAM) is obtained and further processed serve as pseudo label train model fully-supervised In paper, we argue that lost structure information CAM limits its application downstream segmentation, leading deteriorated predictions. Furthermore, inconsistent scores inside same object contradicts common...

Post-training quantization (PTQ) is a popular method for compressing deep neural networks (DNNs) without modifying their original architecture or training procedures. Despite its effectiveness and convenience, the reliability of PTQ methods in presence some extrem cases such as distribution shift data noise remains largely unexplored. This paper first investigates this problem on various commonly-used methods. We aim to answer several research questions related influence calibration set...

Abstract Lyonetia clerkella Linnaeus is one of the notorious pests that are found in orchards. A new and efficient asymmetric synthesis sex pheromone this pest has been developed. Central to approach were 4-(dimethylamino)pyridine N-oxide (DMAPO)/Boc2O-mediated N-acylation, Evans’chiral auxiliaries coupling chiral tosylate with n-propyl magnesium bromide. Furthermore, we have demonstrated scalability acylation.

Abstract Lightning often causes death, injury, and damage to various facilities equipment. Accurately detecting the spatial location of lightning occurrence by predicting thunderstorms is great significance. Traditional detection systems detect measuring sound, light, electromagnetic field information radiated lightning. These methods typically have two problems. First, process signals susceptible interference. Second, equipment cost high not friendly some tasks only targeted at specific...

Neoclytus acuminatus (Fabricius) is a wood-boring pest of forests and orchards.Its contact sex pheromones were (R)-9-methylheptacosane, (R)-7-methylheptacosane (R)-7-methylpentacosane.A new efficient synthesis the pheromone was developed from chiral synthon (S)-2-methyloxirane.This procedure mainly involved ring-opening reaction (S)-2-methyloxirane, SN2 allylation secondary tosylate Wittig coupling aldehyde.

Due to the imbalance between supply and demand, liner container transportation often faces problem of low slot utilization, which will occur in shipping process, such as dry demand exceeding available slots reefer not being fully utilized. This makes it important challenging maintain a balance actual limited number allocated for transport. Therefore, this study proposes flexible allocation method: expanding types containers that can be loaded same slot. method is suitable handling each...

Reliable segmentation of road lines and markings is critical to autonomous driving. Our work motivated by the observations that are (1) frequently occluded in presence moving vehicles, shadow, glare (2) highly structured with low intra-class shape variance overall high appearance consistency. To solve these issues, we propose a Homography Guided Fusion (HomoFusion) module exploit temporally-adjacent video frames for complementary cues facilitating correct classification partially or...

Images are often corrupted by impulse noise due to transmission errors, malfunctioning pixel elements in camera sensors, faulty memory locations the imaging process. This paper proposes a two-phase method for noise. In first phase, suitable is applied identify image pixels contaminated Then, second based upon information on location of noise-free pixels, images recovered using structure decomposition denoising method. order solve model, split bregman iteration combined with alternating...

Abstract The synthesis of a Miltochrista calamine sex pheromone and its diastereomer has been developed. key steps the synthetic approach involved Evans’ chiral auxiliaries addition alkyne to aldehyde, which were firstly applied prepare this diastereomer. could be used trap insects study physiological ecological questions lichen moth.

Point cloud is a widely used geometric data structure in the missions of 3D reconstruction, digital city and geologic survey etc. Extracting sufficient information from point key to deal with aforementioned missions. However, huge number points lead computational complexity inefficiency during training process. To this problem, paper proposes novel framework name Principal Component Analysis Net (PCAPN) for feature extraction cloud. Firstly, sampling module namely Sampling (CPS) designed...

The meteorological landscape is a vital part of tourism viewing and substantial representation conditions, but adequate monitoring recording tools are currently missing. This study proposes an algorithm model based on updated CNN network combined with multi-level feature fusion for identifying typical sceneries. First, five popular datasets created, four image features employed as shallow features: dark channel, brightness, average gradient, contrast. Simultaneously, CNN(Convolutional Neural...

Typical models of lightning detection are based on the acoustic, optical, and electromagnetic field properties lightning, All these require specialized instrumentation to implement. The image-based model is more generalizable scalable than traditional models. just by Camera captures pictures detect lightning. images in a video can be divided into foreground background, understood as video. separated from judging with time index positioning. In order quickly identify video, it necessary...