Wetland Ecotourism Development Using Deep Learning and Grey Clustering Algorithm from the Perspective of Sustainable Development.

The purpose is to promote the sustainable development of wetland ecotourism in China and plan the passenger flow in different tourism periods. This work selects Zhangye Heihe wetland ecotourism spot as the research object. Firstly, the two single wetland ecotourism Demand Prediction Models (DPMs) are proposed based on the time series of the optimized Fuzzy Clustering Algorithm (FCA), grey theory, and the Markov Chain Method. The proposed wetland ecotourism DPM simulates and predicts the ecotourism passenger flow of wetland-scenic spots and verifies the maximum passenger flow. Then, a hybrid model combining the above two single models is proposed, namely, the wetland ecotourism DPM based on an optimized fuzzy grey clustering algorithm. Further, the proposed three models predict the passenger flow in wetland ecotourism spots from 2015 to 2019. A wetland Water Quality Evaluation (WQE) model based on Deep Learning Backpropagation Neural Network (Deep Learning (DL) BPNN) is proposed to evaluate the water quality in different water periods. The results show that the hybrid model's Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) are 1.25% and 0.2532. By comparison, for two single models, the MAPE is 11.67% and 1.45%, respectively, and the RMSE is 0.2526 and 0.1652, respectively. Therefore, the mixed hybrid has the highest accuracy and stability. The water quality of the scenic spot in the wet season is obviously better than that in the dry season and flat season. It is suggested that the natural environmental factors, such as water quality and passenger flow in different periods, should be considered when formulating ecotourism development strategies.

A real-time road detection method based on reorganized lidar data.

Road Detection is a basic task in automated driving field, in which 3D lidar data is commonly used recently. In this paper, we propose to rearrange 3D lidar data into a new organized form to construct direct spatial relationship among point cloud, and put forward new features for real-time road detection tasks. Our model works based on two prerequisites: (1) Road regions are always flatter than non-road regions. (2) Light travels in straight lines in a uniform medium. Based on prerequisite 1, we put forward difference-between-lines feature, while ScanID density and obstacle radial map are generated based on prerequisite 2. According to our method, we construct an array of structures to store and reorganize 3D input firstly. Then, two novel features, difference-between-lines and ScanID density, are extracted, based on which we construct a consistency map and an obstacle map in Bird Eye View (BEV). Finally, the road region is extracted by fusing these two maps and refinement is used to polish up our outcome. We have carried out experiments on the public KITTI-Road benchmark, achieving one of the best performances among the lidar-based road detection methods. To further prove the efficiency of our method on unstructured road, the visual outcomes in rural areas are also proposed.

Multi-appearance segmentation and extended 0-1 programming for dense small object tracking.

Aiming to address dense small object tracking, we propose an image-to-trajectory framework including tracking and detection, where Track-Oriented Multiple Hypothesis Tracking(TOMHT) is revised for tracking. Unlike common cases of multi-object tracking, merged detections and the greater number of objects make dense small object tracking a more challenging problem. Firstly, we handle frequent merged detections through the aspects of detection and hypothesis selection. To tackle merged detection, we revise Local Contrast Method(LCM) and propose a multi-appearance variant, which exploits tree-like topological information and realizes one threshold for one object. Meanwhile, one-to-many constraint is employed via the proposed extended 0-1 programming, which enables hypothesis selection to handle track exclusions caused by merged detections. Secondly, to alleviate the high complexity caused by dense objects, we consider batch optimization and more rigorous and precise pruning technologies. Specifically, we propose autocorrelation based motion score test and two-stage hypotheses pruning. Experimental results are presented to verify the strength of our methods, which indicates speed and performance advantages of our tracker.

Scene text detection via extremal region based double threshold convolutional network classification.

In this paper, we present a robust text detection approach in natural images which is based on region proposal mechanism. A powerful low-level detector named saliency enhanced-MSER extended from the widely-used MSER is proposed by incorporating saliency detection methods, which ensures a high recall rate. Given a natural image, character candidates are extracted from three channels in a perception-based illumination invariant color space by saliency-enhanced MSER algorithm. A discriminative convolutional neural network (CNN) is jointly trained with multi-level information including pixel-level and character-level information as character candidate classifier. Each image patch is classified as strong text, weak text and non-text by double threshold filtering instead of conventional one-step classification, leveraging confident scores obtained via CNN. To further prune non-text regions, we develop a recursive neighborhood search algorithm to track credible texts from weak text set. Finally, characters are grouped into text lines using heuristic features such as spatial location, size, color, and stroke width. We compare our approach with several state-of-the-art methods, and experiments show that our method achieves competitive performance on public datasets ICDAR 2011 and ICDAR 2013.

Evaluation of Pathogenicity of the Fungi Metarhizium anisopliae and Beauveria bassiana in Hazelnut Weevil (Curculio nucum L., Coleoptera, Curculionidae) Larvae.

The nut weevil (Curculio nucum) is one of the most important and widespread pests in hazelnut orchards. In order to screen entomopathogenic fungal strains with high virulence against C. nucum, the growth rate, sporulation, and cumulative mortality of different Metarhizium anisopliae and Beauveria bassiana strains were investigated, and the process by which M. anisopliae CoM 02 infects C. nucum larvae was observed using scanning electron microscopy. The results indicated that the growth rate and sporulation of different fungal strains significantly differed. Thirteen days after inoculation with M. anisopliae CoM 02, the cumulative mortality of C. nucum larvae reached 100 %, which was considerably higher than that of the other five strains. As the most virulent of the six test strains, the cadaver rate, LT50, and LT90 of M. anisopliae CoM 02 were 93.4 %, 7.05 and 11.90 days, respectively. Analysis of the infection process by scanning electron microscopy showed that the spore attachment, hyphal germination, hyphal rapid growth, and sporulation of M. anisopliae CoM 02 occurred on the 3rd, 5th, 7th, and 11th day after inoculation, respectively, indicating that the infection cycle takes approximately 11 days. This finding suggests that the highly virulent M. anisopliae plays an important role in the biocontrol of C. nucum in China.

[Tolerance of human body to simultaneous infusion of 5-fluorouracil and calcium folinate].

BACKGROUND & OBJECTIVE: It has been confirmed that calcium folinate (CF), as a chemotherapy-biochemical modulator of 5-fluorouracil (5-FU), can improve the effect of antitumor of 5-FU. But the usage and dosage of this therapy should be further studied. The purpose of this article was to study the maximum tolerated dose (MTD) and the dose limiting toxicity (DLT) of human body to simultaneously intravenous infusion of 5-FU and CF. METHODS: The cases being up to the selecting standard were treated with DDP+CF/5-FU regimen. The dose of CF and DDP for every course was fixed. The first dose of 5-FU was 500 mg x (m(2) x d)(-1),then the doses were increased by degree of 50 mg x (m(2) x d)(-1), if no severe side effect was observed, up to maximum toxicity tolerated dose. There were 3-6 cases in each degree of dose. RESULTS: Thirty-two cases were treated and given total 64 courses of the administration in all. The overall remission rate was 84.4%. The MTD of 5-FU was 800 mg x (m(2) x d)(-1). Mucositis and diarrhea were found to be the DLT. CONCLUSION: Simultaneous continuous infusion of 5-FU and CF can enhance the dosage of 5-FU, and has a better efficacy to advanced digestive tract, head and neck cancers. So the authors recommend that the dosage of 5-FU in phase II clinical trial is 700 mg x (m(2) x d)(-1). This dose can be used for 5 consecutive days and be repeated after 3 weeks.