Lane detection in driving scenes is an important module for autonomous vehicles and advanced driver assistance systems. In recent years, many sophisticated lane detection methods have been proposed. However, most methods focus on detecting the lane from one single image, and often lead to unsatisfactory performance in handling some extremely-bad situations such as heavy shadow, severe mark degradation, serious vehicle occlusion, and so on. In fact, lanes are continuous line structures on the road. Consequently, the lane that cannot be accurately detected in one current frame may potentially be inferred out by incorporating information of previous frames. To this end, we investigate lane detection by using multiple frames of a continuous driving scene, and propose a hybrid deep architecture by combining the convolutional neural network (CNN) and the recurrent neural network (RNN). Specifically, information of each frame is abstracted by a CNN block, and the CNN features of multiple continuous frames, holding the property of time-series, are then fed into the RNN block for feature learning and lane prediction. Extensive experiments on two large-scale datasets demonstrate that, the proposed method outperforms the competing methods in lane detection, especially in handling difficult situations.
This study investigated the effects of short-term food restriction or supplementation on folliculogenesis and plasma and intrafollicular metabolite and hormone concentrations. Ewes were randomly assigned to three groups: the control group received a maintenance diet (M) while the supplemented group and restricted group received 1.5!M and 0.5!M respectively on days 6-12 of their estrous cycle. Estrus was synchronized by intravaginal progestogen sponges for 12 days. On days 7-12, blood samples were taken. After slaughter, the ovarian follicles were classified and the follicular fluid was collected. Compared with restriction, supplementation shortened the estrous cycle length, decreased the number of follicles 2.5-3.5 mm and follicular fluid estradiol (E 2 ) concentration, increased the number of follicles O3.5 mm and plasma glucose, insulin and glucagon concentrations, and augmented the volume of follicles O2.5 mm. Restricted ewes had higher intrafollicular insulin concentration, but it was similar to that of supplemented ewes. Compared with follicles %2.5 mm, the intrafollicular glucose and E 2 concentrations were increased and the testosterone, insulin, and glucagon concentrations and lactate dehydrogenase (LDH) activity were decreased in follicles O2.5 mm. Only in restricted ewes were intrafollicular LDH and testosterone concentrations in follicles %2.5 mm not different from those in follicles %2.5 mm. In conclusion, the mechanism by which short-term dietary restriction inhibits folliculogenesis may involve responses to intrafollicular increased E 2 , testosterone, and LDH levels in late-stage follicles. This may not be due to the variation of intrafollicular insulin level but rather due to decreased circulating levels of glucose, insulin, and glucagon.
Increasing attention is being devoted to the airborne emissions resulting from a variety of manufacturing processes because of health, safety, and environmental concerns. In this two-part paper, a model is presented for the amount of cutting fluid mist produced by the interaction of the fluid with the rotating cylindrical workpiece during a turning operation. This model is based on relationships that describe cutting fluid atomization, droplet settling, and droplet evaporation. Experiments are performed to validate the model. In Part 1 of the paper, the emphasis is on model development. In the model, thin film theory is used to determine the maximum fluid load that can be supported by a rotating cylindrical workpiece; rotating disk atomization theory is applied to the turning process to predict the mean size of the droplets generated by atomization; and expressions for both the evaporation and settling behavior are established. Droplet size distribution and mass concentration predictions are used to characterize the fluid mist. Model predictions indicate that the droplet mean diameter is affected by both fluid properties and operating conditions, with cutting speed having the most significant affect. Model predictions and experimental results show that the number distribution of droplets within the control volume is dominated by small droplets because of the settling and evaporation phenomena. In Part 2 of the paper, the cutting fluid mist behavior model is validated using the results obtained from a series of experiments.
The integration of intelligent spore image sequence capture device is designed for capturing spores and then generates a series of images. Powdery mildew spore is an extremely harmful bacteria spore in agricultural crops. This paper mainly introduced a new method which could realize the automatic detection of the powdery mildew spores. The new method mainly included pre-processing, image segmentation, feature extraction and identification. Pre-processing included illumination compensation, graying, image enhancement. Image segmentation included binary, image smoothing. Finally identification used BP neural network method. Training pictures used 155 pictures and the correct rate was 95.5%. Testing took 89 pictures, and the correct rate was 63.6%. The result proved the validity and correctness of this method. Key Words-the image sequence capture device; powdery mildew spores image; automatic recognition I.
In Part 1 of this paper a model was developed to describe the formation mechanisms and dynamic behavior of cutting fluid mist. This part of the paper focuses on an experimental investigation of the mist generated by the interaction of the fluid with the rotating cylindrical workpiece during a turning operation and the simulation of the dynamic behavior of the mist droplets, resulting in the prediction of the droplet size distribution and the mass concentration within the machining environment. These simulation results are compared to experimental measurements in order to validate the theoretical model presented in Part 1 of the paper. It is observed that the model predictions accurately characterize the observed experimental behavior.
An electromagnetic flowmeter (EMF) can be used to measure the speed of slurry fluid. In this paper, according to Maxwell's equations and the dB/dt vortex electric field, a step excitation scheme is proposed. The original excitation process of the primary excitation current rise time is divided into a twostep excitation process. Without increasing the excitation current rise time, the equivalent excitation frequency is increased by increasing the primary dB/dt. This method can enhance the EMF's ability to overcome slurry noise. The paper analyzes the design parameters of the step excitation scheme and designs the software and hardware. After that, the paper compares the step excitation scheme with the three-value rectangular wave excitation scheme based on wavelet transform signal processing. Experiments show that the step excitation EMF has a lower steady-state fluctuation rate when measuring mortar and has a stronger ability to overcome slurry noise. Additionally, the step excitation EMF and Yokogawa AXF 040G EMF are compared experimentally. Through flow calibration, clean water measurement, mortar measurement and dynamic response experiment, it is verified that the step excitation EMF has the ability to enhance the resistance to slurry noise at the same excitation frequency.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.