Research of Segmentation Algorithms for Overlapping Chromosomes

Chromosome segmentation is the most important step in the automatic chromosome analysis system. Since in almost every metaphase image partial touching and overlapping of chromosomes are a common phenomenon, how to separate these chromosomes correctly is a difficult yet vital problem. A number of attempts have been made to deal with this problem. This paper is focused on these attempts. Some algorithms are investigated. The principle and the realization of these algorithms are analyzed. Results of these algorithms are compared and discussed.

Multi-segment and Multi-ply Overlapping Process of Multi Coupled Activities Based on Valid Information Evolution

Complex product development will inevitably face the design planning of the multi-coupled activities, and overlapping these activities could potentially reduce product development time, but there is a risk of the additional cost. Although the downstream task information dependence to the upstream task is already considered in the current researches, but the design process overall iteration caused by the information interdependence between activities is hardly discussed; especially the impact on the design process＇ overall iteration from the valid information accumulation process. Secondly, most studies only focus on the single overlapping process of two activities, rarely take multi-segment and multi-ply overlapping process of multi coupled activities into account; especially the inherent link between product development time and cost which originates from the overlapping process of multi coupled activities. For the purpose of solving the above problems, as to the insufficiency of the accumulated valid information in overlapping process, the function of the valid information evolution （VIE） degree is constructed. Stochastic process theory is used to describe the design information exchange and the valid information accumulation in the overlapping segment, and then the planning models of the single overlapping segment are built. On these bases, by analyzing overlapping processes and overlapping features of multi-coupling activities, multi-segment and multi-ply overlapping planning models are built; by sorting overlapping processes and analyzing the construction of these planning models, two conclusions are obtained： （1） As to multi-segment and multi-ply overlapping of multi coupled activities, the total decrement of the task set development time is the sum of the time decrement caused by basic overlapping segments, and minus the sum of the time increment caused by multiple overlapping segments; （2） the total increment of development cost is the sum of the cost increment caused by all overlapping process. And then, based on overlapping degree analysis of these planning models, by the V1E degree function, the four lemmas theory proofs are represented, and two propositions are finally proved： （1） The multi-ply overlapping of the multi coupled activities will weaken the basic overlapping effect on the development cycle time reduction （2） Overlapping the multi coupled activities will decrease product development cycle, but increase product development cost. And there is trade-off between development time and cost. And so, two methods are given to slacken and eliminate multi-ply overlapping effects. At last, an example about a vehicle upper subsystem design illustrates the application of the proposed models; compared with a sequential execution pattern, the decreasing of development cycle （22%） and the increasing of development cost （3%） show the validity of the method in the example The proposed research not only lays a theoretical foundation for correctly planning complex product development process, but also provides specific and effective operation methods for overlapping multi coupled activities.

基于SwinS-YOLACT的番茄采摘机器人实时实例分割算法研究

在设施番茄种植环境中,果实重叠遮挡等情况会影响识别精度。因此,本文提出了一种基于YOLACT的实例分割模型,提高识别精度。首先,对果实重叠遮挡的类别进行细分并增加该类数据集,从而接近真实采摘场景,并在采摘决策中改善重叠遮挡对识别精度的影响;其次,采用Simple Cope-Paste数据增强方法提高了模型的泛化能力,降低了环境因素对实例分割效果的干扰;然后,在YOLACT基础上,引用多尺度特征提取技术克服了单一尺度特征提取的局限性,并降低了模型复杂度;最后,引入Swin Transformer中的Swin-S注意力机制,优化了模型对于番茄实例分割的细节特征提取效果。实验结果表明,本文模型能够一定程度上缓解分割结果中出现的漏检、误检问题,其目标检测平均精度为93.9%,相比于YOLACT、YOLO v8-x、Mask R-CNN、InstaBoost分别提升10.4、4.5、16.3、3.9个百分点;平均分割精度为80.6%,相比于上述模型分别提升4.8、1.5、7.3、4.3个百分点;推理速度为25.6 f/s。该模型综合性能有较强的鲁棒性,兼顾了精度与速度,可为番茄采摘机器人完成视觉任务提供参考。

基于生成对抗网络的手写数字重叠图像分离与识别

为解决手写数字重叠给识别带来的障碍,探索识别重叠手写数字的高效方法,本文提出一种采用生成对抗网络分离重叠手写数字的方法,将重叠手写数字分离成2个单独的数字后再进行识别。分别使用卷积层和反卷积层构建判别器和生成器,充分提取数字特征,减少模型参数量;融入自注意力机制,快速有效提取数字显著区域特征;对生成器和判别器进行谱归一,约束网络梯度;改进模型损失函数,提高生成器分离数字的质量。在通过MNIST数据集构造的数据上进行试验,结果表明:本文提出的方法对重叠手写数字的识别准确率达95.91%;峰值信噪比和结构相似性指数分别为22.11和0.8961,相比CapsNet网络模型有了显著提升。

竖向叠合空间多层桁架分段吊装、高空拼装技术

浙江省全民健身中心工程东西区大跨度钢桁架大部分布置在中间楼层,呈横向隔层布置,其上下部均为常规钢框架结构,边桁架为跃层结构,跨度大、高度大、重量重,根据结构特点和场地情况选择科学合理的施工方案,对工程进度和经济效益影响非常大。

防空导弹IMU状态的时间序列预测方法

惯性测量单元(IMU)是防空导弹的重要部分,其状态数据主要从定期人工测试中获取,效率较低。为了降低对定期测试的依赖,本文通过时间序列预测方法,从已有数据中预测其未来一段时间内的状态。出于小样本考虑,本文使用重叠分段平均进行数据处理,降低数据维度与训练难度,并用长短时记忆网络(LSTM)进行时间维度的预测。本文提出模型在实测数据上进行了验证,在获得最高预测精度的同时保持较低开销。

融合双注意力机制与U-net模型的视网膜血管图像分割方法

视网膜血管分割对于视网膜相关疾病的诊断与治疗有重要临床价值,然而视网膜图像具有复杂性与多样性,处理具有一定难度。为此,针对视网膜血管分割精度不足等问题提出一种融合坐标注意力机制CA、新型无参注意力机制SimAM与基于编码—解码结构的U-net模型的视网膜血管分割方法。该方法通过图片灰度化、限制对比度、直方图均衡化、标准化与gamma校正的预处理操作提高血管与周围组织的对比度;采用重叠平铺Overlap-tile策略将图像分割成子块以增加数据量;在U-net模型的下采样过程中引进坐标注意力机制CA,在上采样过程中引进新型无参注意力机制SimAM,通过训练构建网络模型对测试样本进行分割,得到视网膜血管图像分割结果。该模型在DRIVE数据集上的平均准确率为97.68%,平均灵敏度为80.62%,平均特异性为98.17%,AUC系数为0.9824;同时,与视网膜血管图像分割对照方法相比表现出更佳的分割效果。融合双注意力机制与U-net模型的分割方法可有效提高视网膜血管分割性能。

Feature study for improving Chinese overlapping ambiguity resolution based on SVM

In order to improve Chinese overlapping ambiguity resolution based on a support vector machine, statistical features are studied for representing the feature vectors. First, four statistical parameters-mutual information, accessor variety, two-character word frequency and single-character word frequency are used to describe the feature vectors respectively. Then other parameters are tried to add as complementary features to the parameters which obtain the best results for further improving the classification performance. Experimental results show that features represented by mutual information, single-character word frequency and accessor variety can obtain an optimum result of 94. 39%. Compared with a commonly used word probability model, the accuracy has been improved by 6. 62%. Such comparative results confirm that the classification performance can be improved by feature selection and representation.

A counting method for complex overlapping erythrocytes-based microscopic imaging

Red blood cell(RBC)counting is a standard medical test that can help diagnose various conditions and diseases.Manual counting of blood cells is highly tedious and time consuming.However,new methods for counting blood cells are customary employing both electronic and computer-assisted techniques.Image segmentation is a classical task in most image processing applications which can be used to count blood cells in a microscopic image.In this research work,an approach for erythrocytes counting is proposed.We employed a classification before counting and a new segmentation idea was implemented on the complex overlapping clusters in a microscopic smear image.Experimental results show that the proposed method is of higher counting accuracy and it performs much better than most counting algorithms existed in the situation of three or more RBCs overlapping complexly into a group.The average total erythrocytes counting accuracy of the proposed method reaches 92.9%.

采用边缘曲率进行图像分割的高密度鱼苗计数方法

鱼苗计数是水产养殖中的一个重要课题,在养殖、运输、销售等环节中均需对鱼苗进行定量计数。传统的估算计数方法由于其误差大、效率低等缺陷,已无法满足目前的鱼苗计数需求。为解决高密度、大批量的鱼苗准确计数难题,该研究通过自行搭建的图像采集装置,获取体长1~3 cm的鲢鱼苗在0.25、0.33、0.42、0.50、0.58和0.67尾/cm^(2)等6组不同平面密度下的运动视频,每组鱼苗运动视频共截取3 000帧图像,对所有图像进行预处理,获取鱼苗的二值图像,并提取出图像中的单个连通域及其边缘图像。对含有单尾鱼苗的单个连通域采用直接计数的方法,对含有多尾鱼苗的单个连通域采用先分割后计数的方法。在分割连通域时,一方面计算连通域边缘各处的曲率特征,并标记边缘中的凹点群位置作为预分割位置;另一方面结合连通域的骨架图,根据交叉型、联结型和单凹点群型共3种不同鱼苗重叠类型的形状特征,在连通域中挑选合适的分割位置划取分割线。对鱼苗二值图像中的所有连通域进行依次处理,最后将图像中的连通域总数量作为鱼苗的数量。对18 000张鱼苗图像的计数结果表明,在鱼苗平面密度低于0.58尾/cm^(2)时,鱼苗计数的平均相对误差低于1.37%,计数效率大于100尾/s。该研究解决了高密度鱼苗重叠图像的分割难题,实现了体长1 cm以上鱼苗的准确计数,可为鱼苗计数器的研发提供算法支撑。

基于深度学习的重叠柑橘分割与形态复原

自然场景下柑橘果实的遮挡和重叠现象是自动化采摘面临的一大难题,对此提出了一种基于深度学习的多阶段分割与形态复原方法。首先,通过引入Pointrend分支的Mask R-CNN完成对柑橘的识别及边缘细化的实例分割;其次,基于编解码器结构提出柑橘果实形态粗复原模型,并设计局部惩罚损失函数及交并比形状损失函数;再次,通过机器视觉方法,根据粗复原结果提取感兴趣区域;最后,通过基于部分卷积的形态精复原模型完成重叠柑橘的形态复原。实验结果表明:该方法相比传统方法具有更高的果实定位精度和更好的复原效果,能够为自然场景下的柑橘智能采摘提供参考。

一种基于机器学习的卫星重叠隐蔽通信方法

针对传统卫星重叠通信中单个掩护信号带宽以及功率容限不够的问题,利用卫星转发器频谱环境中多个掩护信号提出了一种频域分割-子谱功率控制联合优化的多掩护信号重叠通信方法,建立了隐蔽通信信号传输性能和隐蔽性能的双目标优化问题,信关站侧采用感知的历史频谱数据训练生成支持向量机回归预测模型,用来预测不同转发器频谱环境下隐蔽信号的通信性能和隐蔽性能,并将训练好的预测模型下载到通信终端;终端侧利用双目标背包算法将支持向量机回归预测模型预测的隐蔽信号的通信性能和隐蔽性能作为价值因素、掩护信号个数作为背包重量来选择转发器频谱环境中的掩护信号,并且求解出隐蔽信号的频域分割和子频谱的功率控制参数,从而实现终端通信信号隐藏在卫星转发器的频谱环境中的目的。

一种基于模板匹配的芯片Frame图像分割算法

针对背景复杂、边界模糊以及芯片相连等特征的芯片Frame图像,提出了一种基于模板匹配的芯片Frame图像分割算法。首先,对整幅芯片Frame图像预分割出多个区域模块;然后,基于区域模块图像采取模板匹配算法匹配单芯片图像;最后,通过合并单芯片的重叠匹配框并记录合并框的坐标信息的方式分割出单芯片图像。实验结果表明:选取合适的模板和阈值,能使该算法的分割准确率达到100%,并且比不基于区域模块匹配的分割算法节省了至少45.76%的分割时间,满足芯片Frame高精度和高速度的分割需求,为芯片图像分割算法的研究提供了一种新思路。

Image segmentation of overlapping leaves based on Chan–Vese model and Sobel operator

To improve the segmentation precision of overlapping crop leaves,this paper presents an effective image segmentation method based on the Chan–Vese model and Sobel operator.The approach consists of three stages.First,a feature that identifies hues with relatively high levels of green is used to extract the region of leaves and remove the background.Second,the Chan–Vese model and improved Sobel operator are implemented to extract the leaf contours and detect the edges,respectively.Third,a target leaf with a complex background and overlapping is extracted by combining the results obtained by the Chan–Vese model and Sobel operator.To verify the effectiveness of the proposed algorithm,a segmentation experiment was performed on 30 images of cucumber leaf.The mean error rate of the proposed method is 0.0428,which is a decrease of 6.54%compared with the mean error rate of the level set method.Experimental results show that the proposed method can accurately extract the target leaf from cucumber leaf images with complex backgrounds and overlapping regions.

Method for segmentation of overlapping fish images in aquaculture

Individual fish segmentation is a prerequisite for feature extraction and object identification in any machine vision system.In this paper,a method for segmentation of overlapping fish images in aquaculture was proposed.First,the shape factor was used to determine whether an overlap exists in the picture.Then,the corner points were extracted using the curvature scale space algorithm,and the skeleton obtained by the improved Zhang-Suen thinning algorithm.Finally,intersecting points were obtained,and the overlapped region was segmented.The results show that the average error rate and average segmentation efficiency of this method was 10%and 90%,respectively.Compared with the traditional watershed method,the separation point is accurate,and the segmentation accuracy is high.Thus,the proposed method achieves better performance in segmentation accuracy and effectiveness.This method can be applied to multi-target segmentation and fish behavior analysis systems,and it can effectively improve recognition precision.

选择性眩光防避系统保护下的司机面部检测

来车发出的强光及低位置太阳光引起的眩目效果会严重危害驾驶安全性,眩光防避系统可智能地在司机的眼睛区域投下阴影,以保护眼睛免受眩目效果的影响.然而,在眼睛被保护性阴影覆盖时,系统仍需对司机脸部进行检测,以实现进一步的监控.针对该问题,采用部分覆膜训练(PMT)、连续亚块训练(CST)以及交叠亚块训练(OST),提出了阴影遮挡下的司机人脸检测算法;同时提出了将司机脸部轮廓检测及头部旋转角度估计方法运用于眩光防避系统,以增强复杂光照条件下的司机脸部检测精度与鲁棒性.实验结果验证了所提算法的有效性.

Segmentation of overlapping cervical nuclei based on the identification

Image segmentation directly determines the performance of automatic screening technique. However,there are overlapping nuclei in nuclei images. It raises a challenge to nuclei segmentation. To solve the problem,a segmentation method of overlapping cervical nuclei based on the identification is proposed. This method consists of three stages： classifier training,recognition and fine segmentation. In the classifier training,feature selection and classifier selection are used to obtain a classifier with high recognition rate. In the recognition,the outputs of the rough segmentation are classified and processed according to their labels. In the fine segmentation,the severely overlapping nuclei are further segmented based on the prior knowledge provided by the recognition. Experiments show that this method can accurately segment overlapping nuclei.

融合K-means与Ncut算法的无遮挡双重叠苹果目标分割与重建

重叠苹果目标的准确分割是采摘机器人必须解决的关键问题之一。针对现有重叠苹果目标分割方法不能保留重叠部分轮廓信息的问题,提出了一种无枝叶遮挡的双果重叠苹果目标分割方法。该方法首先利用K-means聚类算法进行图像分割以提取苹果目标区域,然后利用Normalized Cut(Ncut)算法提取苹果目标轮廓,以实现未被遮挡苹果目标完整轮廓的准确提取,最后利用Spline插值算法对遮挡的苹果目标进行轮廓重建。为了验证算法的有效性,对20幅无枝叶遮挡双果重叠的苹果图像进行试验,并将该算法与寻找2个有效凹点用其连线分割重叠苹果目标,把分离的2个轮廓分别用Hough变换重建轮廓的方法进行对比。试验结果表明,对于图像中未被遮挡的苹果目标,利用该研究算法的平均分割误差为3.15%,提取的苹果目标与原始图像中苹果目标的平均重合度为96.08%,平均误差比Hough变换重建算法低7.73%,平均重合度高9.71%,并且该研究算法能够很好地保留未被遮挡苹果目标的完整轮廓信息,提高了分割精度。对于重叠被遮挡的苹果目标,平均分割误差和平均重合度分别为5.24%和93.81%,比Hough变换重建算法的平均分割误差低11.35%,平均重合度高12.74%,表明该算法可以较好地实现重叠被遮挡苹果目标的轮廓重建,研究结果可为实现枝叶遮挡影响下的多果重叠目标分割与重建提供参考。

基于Snake模型与角点检测的双果重叠苹果目标分割方法

为了实现重叠苹果目标的精确分割,提出了一种Snake模型与角点检测相结合的重叠苹果目标分割方法。该方法首先利用Snake模型得到重叠苹果目标的轮廓,接着采用提出的基于距离测度的角点检测算法寻找重叠苹果目标的角点,针对距离扰动产生伪角点的问题,采用3级db1小波变换得到不含细节信号的近似距离信号,并通过Spline样条内插算法使其恢复到原始信号的长度,从而去除伪角点,最后提出了一种基于长轴分割准则的分割点选取方法并实现了重叠苹果目标的分割。为了验证算法的有效性,利用20幅重叠苹果目标进行了试验,并与人工计算得到的分割线进行了对比,试验结果表明,利用文中算法分割重叠苹果目标的最大误差为13.27°,最小误差为1.20°,平均误差为6.41°,表明Snake模型与角点检测算法相结合对重叠苹果目标具有较好的分割性能,将该方法应用于重叠苹果目标的分割是可行的。

基于凸壳的重叠苹果目标分割与重建算法

重叠苹果目标的分割与定位是影响苹果采摘机器人采摘效率的关键因素之一。为了实现重叠苹果目标的分割与重建,在利用K-means聚类分割算法的基础上,该文提出一种基于凸壳的重叠苹果目标分割方法。通过计算目标凸包与目标相减后的凹区域,将重叠苹果轮廓上的凹点检测转换为凹区域上的凸点检测问题,降低了凹点检测的复杂度。利用相关分割准则实现了凹点匹配并进行目标分割,对分割得到的非完整目标利用Spline插值技术进行目标重建。为了验证算法的有效性,分别利用仿真目标与自然场景下的重叠苹果目标进行了测试,利用该方法得到的苹果目标平均定位误差为14.15%,平均目标重合度为85.08%,表明基于凸壳技术的重叠苹果目标分割方法具有较好的分割性能,将该方法应用于重叠目标分割与重建是有效可行的。