Marine target detection based on Marine-Faster R-CNN for navigation radar plane position indicator images

As a classic deep learning target detection algorithm,Faster R-CNN(region convolutional neural network)has been widely used in high-resolution synthetic aperture radar(SAR)and inverse SAR(ISAR)image detection.However,for most common low-resolution radar plane position indicator(PPI)images,it is difficult to achieve good performance.In this paper,taking navigation radar PPI images as an example,a marine target detection method based on the Marine-Faster R-CNN algorithm is proposed in the case of complex background(e.g.,sea clutter)and target characteristics.The method performs feature extraction and target recognition on PPI images generated by radar echoes with the convolutional neural network(CNN).First,to improve the accuracy of detecting marine targets and reduce the false alarm rate,Faster R-CNN was optimized as the Marine-Faster R-CNN in five respects:new backbone network,anchor size,dense target detection,data sample balance,and scale normalization.Then,JRC(Japan Radio Co.,Ltd.)navigation radar was used to collect echo data under different conditions to build a marine target dataset.Finally,comparisons with the classic Faster R-CNN method and the constant false alarm rate(CFAR)algorithm proved that the proposed method is more accurate and robust,has stronger generalization ability,and can be applied to the detection of marine targets for navigation radar.Its performance was tested with datasets from different observation conditions(sea states,radar parameters,and different targets).

Seasonal Prevalence of <i>Aedes aegypti</i>in Semi-Urban Area of Yangon Region, Myanmar

Prevalence rate of Aedes aegypti was conducted in 20 houses from semi-urban areas of Yangon Region. Larval surveys were done at indoors and outdoors water containers of five types. Prevalence rate of larval density larvae was investigated monthly by standard indices. The highest infestation rate of the container index (CI) was in June 2018 (56.52%), the second highest was in July 2017 (48.36%) and the lowest rate was in April 2017 (5.07%);those of the Jar index (JI) was highest (36.49%) in June and second highest rate (23.8%) was in October 2017. Reasoning the Metal drum (MI) was highest (13.95%) in June 2018 and second highest (6.25%) was in July 2017. The larval infestation rate of Earthen pot (EI) was highest (42.1%) in July 2017. The larval incident rate in almost all indices showed that the highest rate was at the beginning of monsoon season, in June and July, while in the remaining months, the larval incident rate was found to decrease due to the application of insecticides in the study area by the Township Public Health Department. However, the application of insecticides did not cover all the breeding sites of the mosquitoes, the water puddles under their houses were left to apply the insecticides. The positive larval incident rate was assessed by Household (HI), Container index (CI), Breteau index (BI). The highest and second highest positive larval incident rates were all in June 2018 and July 2017 in all indices, HI (27.3% and 23.4%), CI (56.52% and 48.36%), BI (17.56% and 16.79%) and SI (28.49% and 24.38%) respectively. The lowest rate in all indices was 2.56% (HI), 5.07% (IC), 2.67% (BI) and 1.91% (SI) in April. In this study, the fluctuation of indices of infestation rates and positive larval index value was positively correlated in similar trends in the study months. The reason for difficult control measure depends on the water sources under their houses and remains stagnant throughout the year, even in the dry season. High incident and death rates of the children due to Dengue/Dengue Haemorrhagic fever patients in June and July could not be directly correlated with the prevalence of Aedes aegypti. The control measure is needed to wash out the water source under the houses and to apply the insecticides to the all breeding sites.

Pareto Minimizing Total Completion Time and Maximum Cost with Positional Due Indices

In this paper,we study the Pareto optimization scheduling problem on a single machine with positional due indices of jobs to minimize the total completion time and a maximum cost.For this problem,we give two O(n^(4))-time algorithms.