Investigation of cross-sectional image analysis method to determine the blending ratio of polyester/cotton yarn

It has been considered as a great challenge to identify the blending ratio of polyester/cotton yarn in the field of textile industry. A new digital cross-sectional image processing method based on geometrical shape analysis is proposed to improve the measurement accuracy of polyester/cotton blend ratio. A self-developed microscope image capturing system is established to digitalise the cross-sectional images of polyester/cotton blended yarn. One set of image preprocessing algorithm is developed to conduct greyscale inversion, median filtering denoising and binarisation. The specially designed edge detection algorithm is used to identify the continuous profile of fibres. Finally, the roundness value of the cross-sectional fibre is calculated based on the proposed roundness algorithm, it can be used to identify the polyester/cotton fibres and calculate the blending ratio of them. Our experimental results show that the new digital analysis method proposed in this paper is feasible for the measurement of polyester/cotton blended ratio; therefore, it has a good application prospect in the field of textile quality control, including the development of new equipment, methods and standards.     

On-Line Non-Contact System for Grinding Wheel Wear Measurement

An on-line non-contact method for measuring the wear of a form grinding wheel is presented. A CCD (charge coupled device) camera with a selected optical lens and a frame grabber was used to capture the image of a grinding wheel. The analogue signals of the image were transformed into corresponding digital grey level values. Using the binarisation technique, the images of background and the grinding wheel were segmented. Thus the grinding wheel edge was identified. The 'mapping function method' is used to transform an image pixel coordinate to a space coordinate. An auto-focus technology is also developed. The statistics of pixels are used as the focusing index. The signal was sent through an 8255 control card to drive a d.c. motor, and then to control the lens focusing movement to acquire the focal plane. The images before and after the grinding process were captured. The position deviation of the grinding wheel edge was analysed. Then, the grinding wheel wear was evaluated. The wear detection accuracy is about 1 μm.     

Extraction and recognition of artificial text in multimedia documents

Abstract The systems currently available for contentbased image and video retrieval work without semantic knowledge, i. e. they use image processing methods to extract low level features of the data. The similarity obtained by these approaches does not always correspond to the similarity a human user would expect. A way to include more semantic knowledge into the indexing process is to use the text included in the images and video sequences. It is rich in information but easy to use, e. g. by key word based queries. In this paper we present an algorithm to localise artificial text in images and videos using a measure of accumulated gradients and morphological processing. The quality of the localised text is improved by robust multiple frame integration. A new technique for the binarisation of the text boxes based on a criterion maximizing local contrast is proposed. Finally, detection and OCR results for a commercial OCR are presented, justifying the choice of the binarisation technique.An erratum to this article can be found at     

页面扭曲的结构光辅助校正方法

拍摄或扫描图书文档时,所获得的页面图像会有不同程度的扭曲形变,这不仅影响美观或视觉效果,而且影响其深层处理,如OCR(Optical Character Recognition)字符识别。为解决上述问题,提出一种改进的基于模型的扭曲页面校正算法。首先对输入图像进行转正预处理,并通过以图像梯度信息确定阈值的二值化方法去除页面的灰色背景,然后利用简易的直线结构光提取页面文字行点集,由点集中心点曲线构建柱面模型进行页面校正。实验表明该方法能适应更多不同的页面扭曲类型,校正和去背景处理效果好、效率高,可显著提高OCR识别率,而且系统结构简便,容易实现。     

A neuro-fuzzy technique for document binarisation

This paper proposes a new neuro-fuzzy technique suitable for binarisation or, in general, the colour reduction of digital documents. The proposed approach uses the image colour values and additional local spatial features extracted in the neighbourhood of the pixels. Both image and local features values feed a Kohonen self-organised feature map (SOFM) neural network classifier. After training, the neurons of the output competition layer of the SOFM define a first approach of the final classes. Using the content of these classes, fuzzy membership functions are obtained that are next used by the fuzzy C-means (FCM) algorithm in order to obtain the colours of the final document. The method can be applied to greyscale and colour documents; it is suitable for improving blurring and badly illuminated documents and can be easily modified to accommodate any type of spatial characteristics.