基于卷积神经网络马铃薯叶片病害识别和病斑检测*

针对自然环境下马铃薯叶片病害识别率低和晚疫病斑定位难的问题,基于大田环境中采集的马铃薯叶片图像,首先对马铃薯叶片病害进行识别,对比AlexNet、VGG16、InceptionV3、ResNet50、MobileNet五种神经网络模型,结果表明InceptionV3模型的识别效果准确率最高,可达98.00%。其次对马铃薯叶片的晚疫病斑进行检测,提出一种改进型的CenterNet-SPP模型,该模型通过特征提取网络获取对象的中心点,再通过中心点回归获得中心点偏移量、目标大小等图像信息,训练后的模型在验证集下的mAP可达90.03%,以F₁为评价值分析对比其它目标检测模型,CenterNet-SPP模型的效果最好,准确率为94.93%,召回率为90.34%,F₁值为92.58%,平均检测一张图像耗时0.10 s。为自然环境下马铃薯叶片病害识别和检测提供较为全面的深度学习算法和模型研究基础。     

Turbidity‐induced changes in emergent effects of multiple predators with different foraging strategies

Abstract – In natural systems, prey frequently interact with multiple predators and the outcome often cannot be predicted by summing the effects of individual predator species. Multiple predator interactions can create emergent effects for prey, but how those change across environmental gradients is poorly understood. Turbidity is an environmental factor in aquatic systems that may influence multiple predator effects on prey. Interactions between a cruising predator (largemouth bass Micropterus salmoides) and an ambush predator (muskellunge Esox masquinongy) and their combination foraging on a shared prey (bluegill Lepomis macrochirus) were examined across a turbidity gradient. Turbidity modified multiple predator effects on prey. In clear water, combined predators consumed in total more prey than expected from individual predator treatments, suggesting risk enhancement for prey. In moderately turbid water, the predators consumed fewer prey together than expected, suggesting a risk reduction for prey. At high turbidity, there were no apparent emergent effects; however, the cruising predator consumed more prey than the ambush predator, suggesting an advantage for this predator. Understanding multiple predator traits across a gradient of turbidity increases our understanding of how complex natural systems function.     

Object‐Based Mapping of Karst Rocky Desertification using a Support Vector Machine

Accurate and cost‐effective mapping of karst rocky desertification (KRD) is still a challenge at the regional and national scale. Visual interpretation has been utilised in the majority of studies, while an automated method based on pixel data has been investigated repeatedly. An object‐based method coupling with support vector machine (SVM) was developed and tested using Enhanced Thematic Mapper Plus (ETM+) images from three selected counties (Liujiang, Changshun and Zhenyuan) with different karst landscapes in SW China. The method supports a strategy of defining a mapping unit. It combined ETM+ images and ancillary data including elevation, slope and Normalized Difference Vegetation Index images. A sequence of scale parameters estimation, image segmentation, training data sampling, SVM parameters tuning and object classification was performed to achieve the mapping. A quantitative and semi‐automated approach was used to estimate scale parameters for segmenting an object at an optimal scale. We calculated the sum of area‐weighted standard deviation (WS), rate of change for WS, local variance (LV) and rate of change for LV at each scale level, and the threshold of the aforementioned index that indicated the optimal segment level and merge level. The KRD classification results had overall accuracies of 85·50, 84·00 and 84·86 per cent for Liujiang, Changshun and Zhenyuan, respectively, and kappa coefficients are up to 0·8062, 0·7917 and 0·8083, respectively. This approach mapped six classes of KRD and offered a visually appealing presentation. Moreover, it proposed a conceptual and size‐variable object from the classification standard of KRD. The results demonstrate that the application of our method provides an efficient approach for the mapping of KRD. Copyright © 2012 John Wiley & Sons, Ltd.     

基于高斯自适应拟合的苹果目标分割方法研究

苹果目标的准确识别是苹果机械化采摘需要解决的关键问题之一。为此,基于YUV颜色空间模型,提出了一种结合色差分量与高斯自适应拟合算法的苹果目标分割方法。该方法采用首先将苹果目标由RGB颜色空间转换至YUV颜色空间,并利用色差分量V建立果实与背景分割的高斯分布拟合模型,根据拟合结果自动获取分割阈值,以实现自然场景下苹果目标的准确分割。为了验证文中算法的有效性,利用多幅图像进行了试验并与Otsu自适应阈值分割算法进行了比较。试验结果表明,采用文中算法得到的苹果果实的平均检出率达87.08%,识别率领先Otsu算法9.91%。因此,对于着色度较为均匀的苹果目标,采用高斯自适应拟合方法可以有效提高其识别率。     

INVITED REVIEW—IMAGE REGISTRATION IN VETERINARY RADIATION ONCOLOGY: INDICATIONS,IMPLICATIONS, AND FUTURE ADVANCES

The field of veterinary radiation therapy (RT) has gained substantial momentum in recent decades with significant advances in conformal treatment planning, image‐guided radiation therapy (IGRT), and intensity‐modulated (IMRT) techniques. At the root of these advancements lie improvements in tumor imaging, image alignment (registration), target volume delineation, and identification of critical structures. Image registration has been widely used to combine information from multimodality images such as computerized tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) to improve the accuracy of radiation delivery and reliably identify tumor‐bearing areas. Many different techniques have been applied in image registration. This review provides an overview of medical image registration in RT and its applications in veterinary oncology. A summary of the most commonly used approaches in human and veterinary medicine is presented along with their current use in IGRT and adaptive radiation therapy (ART). It is important to realize that registration does not guarantee that target volumes, such as the gross tumor volume (GTV), are correctly identified on the image being registered, as limitations unique to registration algorithms exist. Research involving novel registration frameworks for automatic segmentation of tumor volumes is ongoing and comparative oncology programs offer a unique opportunity to test the efficacy of proposed algorithms.     

基于"3S"技术的甘肃子午岭自然保护区土地利用变化研究

以甘肃省子午岭自然保护区为研究对象,运用"3S"技术,对研究区的地形图、不同时相的TM遥感影像进行数据预处理,并分类提取了研究区2个时段的土地利用信息.对分类信息进行统计分析后并与实地调查数据进行比较,结果表明,近20年来甘肃子午岭自然保护区的森林覆盖面积在稳步提高,乔木林面积明显增加.为甘肃子午岭自然保护区工程的建设提供了基础信息,为林区植物种群生长及分布调查研究提供了宏观数据,并且输出了研究区不同时相土地利用专题图.     

Practical Method of High-Speed Access for Medical Digital Image Sequences

The high-speed access for medical digital image sequences was restricted to special device and complicated system in the past. The authors present a new method of high-speed access for medical digital image sequences which can be performed on general computer by means of the combination of Serial ATA(SATA) and Redundant Array of Independent Disk(RAID).Then they introduce the scheme of system configuration and software program. Lastly the testing result is given and the speediness,cheapness and conveniency of this method are elucidated.     

Visualization of the FEM Analysis Data of 3D on the Cut Plane

The colored patterns on arbitrary cut plane is an important method in 3D data-field visualization of finite element.By using the technology of the Object-Oriented Programming and the method of scan-line,the colored patterns on arbitrary cut plane are given.The example by the OPENGL of Visual C++ R6 can support the theory satisfactorily.