共查询到17条相似文献,搜索用时 50 毫秒
1.
眼底图像中的视盘在青光眼筛查和诊断中起着重要作用。因此,从眼底图像中对视盘进行准确、快速地定位与分割具有重要意义。在过去,研究者们已经进行了对视盘的深入研究,但如何提高定位准确率和分割精度仍是视盘分割的一大难题。对此本文提出一种采用深度学习结构U-Net的视网膜视盘自动分割的方法,该方法结合机器学习,通过深度网络提取输入图像的视盘特征,从而得出相应的分割结果图。相对于传统的视盘分割方法,本文的U-Net神经网络能够有效学习有利于分割视盘的特征,从而提高分割的精确度,而且分割耗时更短。 相似文献
2.
彩色眼底图像视盘自动定位与分割 总被引:1,自引:0,他引:1
针对彩色眼底图像视盘定位时图像边缘高亮环对定位准确率的影响,提出了一种有效的图像预处理方法。针对已有的视盘分割算法中存在的问题,提出了一种结合形态学、椭圆拟合及梯度矢量流(GVF)Snake模型的分割算法。提出的预处理方法首先利用最小二乘法拟合出眼底图像的边界,然后裁剪掉边界的一部分高亮像素点,最后进行视盘定位。视盘分割算法则首先进行血管擦除,然后用椭圆拟合提取初始轮廓,最后使用GVFSnake精确调整视盘边界。用提出的方法对Messidor眼底图像数据库1200幅图像上进行了实验,结果显示:视盘定位准确率由原来没经过预处理的95.4%提升到了98.7%;视盘分割错误率与当前已知最好的算法相比由12.5%降低到了9.39%。结果表明:提出的眼底图像视盘自动定位与分割方法准确率高、实用性强,可以用于眼科疾病的计算机辅助诊断。 相似文献
3.
针对当前深度神经网络模型在检测小缺陷目标时性能较差的问题,提出了一种基于改进U-Net的工件表面缺陷分割方法。该方法设计了一种仅下采样3次的U型网络,在保持图像特征分辨率的同时获得足够的感受野,有效解决神经网络多次下采样造成的小目标信息丢失问题;引入Dice损失和Focal损失组成的混合损失函数,通过增强分割损失权重并抑制背景信息来提高分割效果,有效解决小缺陷目标的低概率密度问题。通过在表面缺陷数据集上的大量实验和分析,结果表明该算法能够很好地细分出缺陷区域,并在分割精度与速度之间获得平衡。 相似文献
4.
在面对视网膜图像中细小血管时,现有算法存在分割精度低的问题。通过在U-Net中引入残差模块与细节增强注意力机制模块,提出了一种改进的U-Net分割算法。在编解码阶段,用残差模块取代传统的卷积模块,解决了网络随深度增加而退化的问题;同时在编码器和解码器间增加细节增强注意力机制,减少编码器输出中的无用信息,从而提高网络抓取有效特征信息的敏感度。此外,基于标准图像集DRIVE的实验结果表明,所提算法的分割准确率、灵敏度与F1值较U-Net算法分别提高了0.46%,2.14%,1.56%,优于传统分割算法。 相似文献
5.
基于改进的Hough变换图像分割方法 总被引:39,自引:10,他引:29
在图像处理和计算机视觉中,Hough变换是一种应用非常广泛的图像边缘检测技术,针对传统Hough变换算法所需的存储容量大、计算量大、速度慢、效率低,不能确定曲线端点及长度的问题,提出了一种改进的Hough变换方法,它根据Hough变换思想的逆变换,采用对参数空间逐步细分的方法,逐步排除不包含直线的区域,能够有效地减少存储容量,提高运行效率,并能有效地求出曲线轮廓的端点及长度.该方法已成功运用于车牌自动识别系统的车牌分割中. 相似文献
6.
7.
针对现有基于深度学习的图像分割算法在球栅阵列(BGA)焊点气泡检测中检测效率较低,无法满足工业生产中实时性的检测需求,提出了一种基于改进U-Net的球栅阵列缺陷识别方法。该方法在现有的U-Net经典网络的基础上提出用深度可分离卷积与密集连接结合的轻量密集连接单元替换常规的卷积单元,同时添加多尺度跳跃连接减少编解码特征之间的差异,实现针对BGA焊点气泡的精确分割和提取。采用自建数据集对该方法的有效性进行实验,结果表明,改进的U-Net模型网络在减少U-Net网络计算复杂度的同时提升了网络性能,能够增加BGA焊点气泡的检测效率。 相似文献
8.
9.
基于曲线演化的水平集算法近来已被广泛应用于医学图像分割中,根据分割医学图像的鲁棒性实时性的要求,提出一种新的基于改进窄带法的图像分割方法INBM(Improved narrow band method).INBM首先将均匀采样的图像映射到对数极坐标系中,由视网膜空间分辨率机制可知,注视点都在图像兴趣区,由此形成初始轮廓,然后用改进的窄带水平集(Level Set)方法演化曲线得到最终分割结果.改进窄带法是通过降低窄带区域内的水平集函数求解个数,来减少计算时间.实验结果表明,该方法能够快速、准确地得到医学图像的结果. 相似文献
10.
大型锻件是国家重型制造装备和重大工程建设所必需的关键和核心基础部件,目前大多还处于人工检测的阶段,如环型锻件的锻造检测过程中,通常需要人工执尺子测量,存在人为的判断误差。另一方面,锻件和所处环境温度高且测量速度慢。使用深度学习与图像处理方式测量环形锻件的直径,首先,由U-Net语义分割网络分割出锻件有效部分,其次,将分割出的图像从RGB空间转换到HSV彩色空间并调整阈值,最后,由弧支撑线段来拟合锻件,实现锻件的直径在线检测。该方法能去除复杂背景,降低锻件高温高亮导致边缘模糊的影响,减少误检和漏检且检测速度较快。实验结果表明,锻件外环和内环检测的平均绝对误差约为2.77cm,2.02cm以及平均的检测总时间约为0.4s,在环形锻件检测领域具有借鉴意义。 相似文献
11.
针对现有方法分割弱边缘铸件CT图像难度大、精度低、鲁棒性差的问题,提出一种融合残差模块与混合注意力机制的U型网络分割算法(AttRes-U-Nets)。该算法以U-Net网络为基础,首先构建深度残差网络ResNets作为算法的编码网络,解决传统U-Net网络特征提取能力不足的问题;然后,引入改进后的混合注意力机制,突出分割目标区域与通道的特征响应,提高网络灵敏度;最后,将Focal loss与Dice loss结合为一种新损失函数FD loss缓解样本不平衡带来的负面影响。使用120阀体数据集对算法性能进行验证,实验结果表明,本文算法对铸件分割的像素准确率(PA)和交互比(IoU)分别达到98.72%和97.40%,优于传统U-Net算法与其他主流语义分割算法,为弱边缘分割提供了新思路。 相似文献
12.
Sara M. M. El-Desoky;Ruwaida Elhanbaly;Abdalla Hifny;Nagwa Ibrahim;Wafaa Gaber; 《Microscopy research and technique》2024,87(4):774-789
The retina consists of various cell types arranged in eight cell layers and two membranes that originate from the neuroectodermal cells. In this study, the timing of differentiation and distribution of the cellular components and the layers of the rabbit retina are investigated using light and electron microscopy and immunohistochemical techniques. There were 32 rabbit embryos and 12 rabbits used. The rabbit retina begins its prenatal development on the 10th day of gestation in the form of optic cup. The process of neuro- and gliogenesis occurs in several stages: In the first stage, the ganglionic cells are differentiated at the 15th day. The second stage includes the differentiation of Muller, amacrine, and cone cells on the 23rd day. The differentiation of bipolar, horizontal, and rod cells and formation of the inner segments of the photoreceptors consider the late stage that occurs by the 27th and 30th day of gestation. On the first week of age postnatally, the outer segments of the photoreceptors are developed. S100 protein is expressed by the Muller cells and its processes that traverse the retina from the outer to the inner limiting membranes. Calretinin is intensely labeled within the amacrine and displaced amacrine cells. Ganglionic cells exhibited moderate immunoreactivity for calretinin confined to their cytoplasm and dendrites. In conclusion, all stages of neuro- and gliogenesis of the rabbit retina occur during the embryonic period. Then, the retina continues its development postnatally by formation of the photoreceptor outer segments and all layers of the retina become established. 相似文献
13.
Hidayat Ullah Tanzila Saba Naveed Islam Naveed Abbas Amjad Rehman Zahid Mehmood Adeel Anjum 《Microscopy research and technique》2019,82(4):361-372
Atomic recognition of the Exudates (EXs), the major symbol of diabetic retinopathy is essential for automated retinal images analysis. In this article, we proposed a novel machine learning technique for early detection and classification of EXs in color fundus images. The major challenge observed in the classification technique is the selection of optimal features to reduce computational time and space complexity and to provide a high degree of classification accuracy. To address these challenges, this article proposed an evolutionary algorithm based solution for optimal feature selection, which accelerates the classification process and reduces computational complexity. Similarly, three well‐known classifiers that is, Naïve Bayes classifier, Support Vector Machine, and Artificial Neural Network are used for the classification of EXs. Moreover, an ensemble‐based classifier is used for the selection of best classifier on the basis of majority voting technique. Experiments are performed on three well‐known benchmark datasets and a real dataset developed at local Hospital. It has been observed that the proposed technique achieved an accuracy of 98% in the detection and classification of EXs in color fundus images. 相似文献
14.
Shahzad Akbar Syed Ale Hassan Ayesha Shoukat Jaber Alyami Saeed Ali Bahaj 《Microscopy research and technique》2022,85(6):2259-2276
Glaucoma disease in humans can lead to blindness if it progresses to the point where it affects the oculus' optic nerve head. It is not easily detected since there are no symptoms, but it can be detected using tonometry, ophthalmoscopy, and perimeter. However, advances in artificial intelligence approaches have permitted machine learning techniques to diagnose at an early stage. Numerous methods have been proposed using Machine Learning to diagnose glaucoma with different data sets and techniques but these are complex methods. Although, medical imaging instruments are used as glaucoma screening methods, fundus imaging specifically is the most used screening technique for glaucoma detection. This study presents a novel DenseNet and DarkNet combination to classify normal and glaucoma affected fundus image. These frameworks have been trained and tested on three data sets of high-resolution fundus (HRF), RIM 1, and ACRIMA. A total of 658 images have been used for healthy eyes and 612 images for glaucoma-affected eyes classification. It has also been observed that the fusion of DenseNet and DarkNet outperforms the two CNN networks and achieved 99.7% accuracy, 98.9% sensitivity, 100% specificity for the HRF database. In contrast, for the RIM1 database, 89.3% accuracy, 93.3% sensitivity, 88.46% specificity has been attained. Moreover, for the ACRIMA database, 99% accuracy, 100% sensitivity, 99% specificity has been achieved. Therefore, the proposed method is robust and efficient with less computational time and complexity compared to the literature available. 相似文献
15.
16.
在全自动纸杯成型机的设计开发过程中,需要进行大量的计算分析,其中凸轮分度器是其实现全自动化的核心部件,本文结合实例就选择合理型号的凸轮分度器的全过程作了介绍. 相似文献
17.