基于脑电信号评价情绪状态及听觉激励对改善情绪的研究

如何从脑活动数据中发现具有生物学意义的知识及规律,正成为当今神经信息学理论与实践研究的热点与难点.基于脑电的情绪识别和情绪及其脑机制的关联研究已经成为神经工程和生物医学工程领域的热门课题.本研究基于脑电信号评价情绪状态,尝试借助脑电耳机来监测"毫无头绪"的脑电波,重点开展了听觉激励对改善焦虑情绪状态的实验研究.借助特定中国古风纯音乐产生听觉信号激励,激发出被试者平静放松的情绪状态,从而改善焦虑,发现音乐和情绪之间可能存在的关系.将来在BCI技术研究的基础之上,情绪识别可以进一步满足患有特定疾病（ALS症、脑瘫、脑干中风、脊髓损伤等）的患者人群的某些更高级的需求,即通过脑电分析出其对事物的喜好度,比如音乐的选择,从而设计出更加友善和智能的BCI系统,在医疗康复领域具有很好的应用前景.     

基于镜像图的LRC和CRC偏差结合的人脸识别

为了提高人脸识别率及更好地显示人脸特征,本文提出了一种基于镜像图的LRC和CRC偏差结合的人脸识别方法.该方法首先生成一种镜像人脸,再通过融合原始人脸和镜像人脸形成新的混合训练样本,最后利用LRC和CRC偏差结合进行人脸识别.新方法增加了训练样本的数目,克服了由于光照和姿态等外部因素带来的影响.实验结果表明,镜像图与LRC和CRC偏差结合的人脸识别方法提高了人脸识别的准确性.     

基于图像处理的自动野外救援系统设计

针对野外救援的不及时性,结合人工智能等技术的应用,提出了一种新的自动野外救援系统.人类在野外和海上活动中受困后,除了依靠自身能力和装备进行求生之外,外部的有效及时救援是提高生存几率的关键.自动野外救援系统,由自动报警装置、手动报警装置实现报警,触发整个系统的运行.自动救援装置可以通过定位模块实现定位,通过执行器移动到随身装置附近实现救援,随身装置还可以发出信标导引信号;自动救援装置还可以通过传感器检测信标,通过相位检测技术等计算待救者的坐标并规划路径,实现更精准的定位与救援.     

基于机器学习的重庆长江航道雾图像特征识别研究

基于重庆市境内长江航道雷达站拍摄的雾天气过程影像资料,利用K最近邻、支持向量机、BP神经网络、随机森林等机器学习算法,对无雾和5类有雾天气个例进行图像识别训练,构建雾图像识别模型,并检验了识别准确率。结果表明：机器学习能够有效识别雾图像,随机森林算法的识别效果优于其余3种算法。对于能见度超过1500 m的无雾天气,模型的识别准确率为100%,对于能见度在1000—1500 m范围内的轻雾、能见度低于50 m的强浓雾,模型的识别准确率在90%以上,对于能见度在50—1000 m范围内的雾、大雾和浓雾,识别准确率超过70%。     

科尔沁湿草甸参考作物蒸散发模拟分析

为探明气候变化下干旱半干旱地区湿草甸参考作物蒸散发(ET₀)影响因子,使用FAO 56 P-M模型对科尔沁湿草甸ET₀进行模拟,利用涡度相关系统对模型的适用性进行评价,并通过通径分析及指标敏感性分析对ET₀的影响因子进行辨识。结果表明:(1)小时尺度模拟精度最高,日尺度次之,月尺度较差,小时尺度上晴、阴、雨3种天气条件下模拟效果不同,晴天最优,阴雨天较差。(2)ET₀年内变化呈单峰曲线状,生长季明显高于非生长季,集中在3—10月,占全年89.79%。生长季典型晴天ET₀逐小时分布特征遵循倒“U”单峰型变化规律。(3)通径分析结果显示,对ET₀的通径系数以及对回归方程估测可靠程度E的总贡献均表现为VPD(饱和水汽压差) > T_min(最低气温) > R_n(冠层表面净辐射)>u₂(2 m高度风速),即VPD为影响ET₀最重要的因子;指标敏感性分析中,在去除VPD后引起的E变化最大,说明ET₀对VPD的变化最为敏感,其次为u₂、T_min和R_n。     

Statistical characterization of frost zones: Case of tea freeze damage in the Kenyan highlands

Frequent frost occurrences in the Kenyan highlands have had devastating effects on agricultural productivity. With inadequate management systems to mitigate the impacts, farmers have often had to bear the burden of losses resulting from frost damage. While agriculture in Kenya remains dependent on weather and climate, the agricultural economy of Kenya continues to suffer, underscoring the need for building local knowledge as basis for development of early warning systems. The current paper attempts to delineate frost zones by statistically characterizing them based on known risk factors related to topography (elevation, convexity, aspect, upslope flow length) and Land Surface Temperature (LST) derived from Moderate Resolution Imaging Spectroradiometer (MODIS). Through binary logistic regression, a logistic regression model was developed utilizing observation data (frost occurrence and non-occurrence) as a binary dependent variable to estimate the probability of frost occurrence. Assuming a 0.5 probability cut-off threshold between frost occurrence and non-occurrence, an overall accuracy of 81% with area under Receiver Operating Characteristics (ROC) Curve of 0.88 was obtained. No evidence of lack of model fit was detected. This model outperforms the currently operational model that utilizes MODIS LST alone to detect frost zones in the Kenyan tea plantations. It provides an improved method for effective delineation of frost zones by incorporating local topographic characteristics.     

Winter cover crops in Dutch maize fields: Variability in quality and its drivers assessed from multi-temporal Sentinel-2 imagery

Planting a cover crop between the main cropping seasons is an agricultural management measure with multiple potential benefits for sustainable food production. In the maize production system of the Netherlands, an effective establishment of a winter cover crop is important for reducing nitrogen leaching to groundwater. Cover crop establishment after maize cultivation is obliged by law for sandy soils and consequently implemented on nearly all maize fields, but the winter-time vegetative ground cover varies significantly between fields. The objectives of this study are to assess the variability in winter vegetative cover and evaluate to what extent this variability can be explained by the timing of cover crop establishment and weather conditions in two growing seasons (2017–2018). We used Sentinel-2 satellite imagery to construct NDVI time series for fields known to be cultivated with maize within the province of Overijssel. We fitted piecewise logistic functions to the time series in order to estimate cover crop sowing date and retrieve the fitted NDVI value for 1 December (NDVI_Dec). We used NDVI_Dec to represent the quality of cover crop establishment at the start of the winter season. The Sentinel-2 estimated sowing dates compared reasonably with ground reference data for eight fields (RMSE = 6.6 days). The two analysed years differed considerably, with 2018 being much drier and warmer during summer. This drought resulted in an earlier estimated cover crop sowing date (on average 19 days) and an NDVI_Dec value that was 0.2 higher than in 2017. Combining both years and all fields, we found that Sentinel-2 retrieved sowing dates could explain 55% of the NDVI_Dec variability. This corresponded to a positive relationship (R² = 0.50) between NDVI_Dec and the cumulative growing degree days (GDD) between sowing date and 1 December until reaching 400 GDD. Based on cumulative GDD derived from two weather stations within Overijssel, we found that on average for the past three decades a sowing date of 19 September (± 7 days) allowed to attain these 400 GDD; this provides support for the current legislation that states that from 2019 onwards a cover crop should be sown before 1 October. To meet this deadline, while simultaneously ascertaining a harvest-ready main crop, in practice implies that undersowing of the cover crop during spring will gain importance. Our results show that Sentinel-2 NDVI time series can assess the effectiveness and timing of cover crop growth for small agricultural fields, and as such has potential to inform regulatory frameworks as well as farmers with actionable information that may help to reduce nitrogen leaching.     

规则池塘群的复杂直线模式识别

针对现有面要素直线模式识别方法难以解决规则池塘中破碎区域的问题,提出了一种规则池塘群的复杂直线模式识别方法。首先,分析了规则池塘群的空间特征和认知特点,提出了“主次关系→并列关系→直线模式→复杂直线模式”的多层次认知顺序;其次,设计了主次关系池塘组、并列关系池塘组、直线模式池塘群的识别方法;最后,构建了复杂直线模式池塘群的识别模型。试验表明:本文方法能够消除破碎区域对规则池塘群直线模式提取的不利影响,有效提升复杂直线模式识别的质量。     

复杂卫星图像中的小目标船舶识别

船舶作为海上的重要目标,实现对船舶自动识别有重要的意义。针对卫星图像中云雾、海岸背景等复杂海情对船舶识别带来的干扰,以及小目标船舶高漏检率问题,本文提出一种多尺度深度学习模型训练策略,在此基础上构建了一种船舶识别的深度学习网络,该网络可分为多尺度训练、特征提取、生成目标建议区域、船舶分类这4个部分。首先,采用多尺度的训练策略,将多尺度的船舶样本送入网络中进行训练,这样在训练样本中加入了大量小目标船舶的样本,使网络充分提取到小目标船舶的特征;其次,通过卷积神经网络对目标船舶进行特征自适应提取;然后,目标区域建议网络可依据卷积神经网络提取到的特征,在图像中找到感兴趣目标区域,即框定船舶的位置;最后,通过多个全连接层的组合,将高维特征映射到一个4元组中,再运用分类函数输出每一类船舶的概率值,概率值最大的则为该船舶的类别。同时为解决云雾遮挡和海岸背景的干扰,采用了一种负样本增强学习的方法,在样本数据集中加入了大量只含有云雾和海岸背景的图片,进行负样本扩充,增强网络模型对云雾及海岸背景的特征学习能力,以此解决复杂海情的影响。实验结果表明,所提方法有效解决了复杂海情条件下的船舶识别难,以及小目标船舶识别难的问题,实现了复杂海情条件下的船舶识别。同时,与现有成熟的深度学习目标识别算法相比,本文算法的精确度和召回率分别提升了6.98%和18.17%,所训练的模型具有良好的泛化能力和鲁棒性。     

基于计算机视觉的头足类角质颚特征研究Ⅰ:轮廓与特征点提取

轮廓与特征点研究是头足类角质颚形态特征鉴别的基本方法,对于轮廓与特征点的提取最常用的方法是手动描绘与标定,利用计算机视觉进行轮廓与特征点的提取,不仅可以降低手动提取带来的误差、提高准确性,而且更加快速、便捷。文章将利用计算机视觉提取头足类角质颚的轮廓与特征点,首先将自制装置拍摄得到的角质颚三视图放入MTALAB软件中进行编程处理,然后利用Canny算法提取角质颚轮廓,最后根据地标点的定义标定特征点位置并建立空间坐标系得到角质颚的特征点坐标。研究结果显示,利用计算机视觉提取角质颚的轮廓图像以及特征点坐标是可行的,当标准差σ取值为0.1时角质颚轮廓图效果最佳,在得到的轮廓图上进行特征点的标定,通过迭代遍历轮廓图获得各个特征点的空间坐标。研究分析认为,将计算机视觉应用于头足类角质颚形态学的研究可以提高研究的便捷性,同时也为后续的研究提供了新的实验思路和方法。