Application of Image Edge Detection Technology in Welding Environment Recognition (2)

2.1.2 Laplacian operator The Laplacian operator is a second-order differential operator whose general representation in digital images is:

Where s is a set of neighbors centered on f(m,n), but 4 or 8 neighbors. Can be directly As edge pixel grayscale, you can also The pixels act as edges.

Gradient operators and Laplacian operators are sensitive to noise. In this case, the neighborhood averaging method can be used for smoothing before the edge extraction, and the Gaussian two-dimensional low-pass filter can be used to filter the image, and then Laplacian edge extraction. This is the commonly used Laplacian-Gauss operator. There are many noise points on the welded workpiece, and the differential operation has a "diffusion" effect on those isolated noise points, especially the Laplacian operator, which not only spreads but also increases the strength significantly. So it is best to remove the noise before the differential operator detects the edge. For butt welds, when the gap is small, the edge features are shown as thin lines, which become wider after the differential operation. Since the gradient operator can detect the direction of the edge of the image, it is more suitable for the identification of the welding environment. 2.2 Wavelet multi-scale edge detection method [9][10]

Wavelet transform can be understood as a common understanding: compare the original signal with the left end of the wavelet function, find the similarity coefficient of the two functions, and then shift the wavelet function to the right of a wavelet function for comparison and calculation until the whole is completed. The operation of the signal; this gives the wavelet coefficients at a scale. The wavelet function is expanded, and the above process is repeated to obtain wavelet coefficients at a series of scales. The point of abrupt changes in the image is a key feature in the analysis of the image, usually the edge feature of interest. Edge detection is to find the sudden change point of the step from the gradient direction of the wavelet coefficient. In order to detect the edge and detail features of the large target structure in the image, the researchers proposed the concept of multi-scale edge detection, that is, detecting the large edge of the target on a large scale and detecting the target details on a small scale. Related theories can be found in the literature [9] [10]. This method is one of the hot spots in current image processing and has a good development prospect. It has been applied to the processing of weld pool images [10].

For the welding environment, the method has good adaptability, and the workpiece or the molten pool can be searched for the target from a large scale, and then the details of interest are extracted.

2.3 Mathematical Morphology Method [11]

Mathematical morphology is the theory of studying the morphological structure features and fast parallel processing methods of digital images. It is the purpose of structural analysis and feature extraction by morphological transformation of target images. Mathematical morphology takes the morphological features of images as the research object. Its main content is to design a set of concepts, transformations and algorithms to describe the basic features and basic structure of the image, that is, to describe the elements and elements, parts and parts of the image. Relationship. The object and image features in the image are directly dependent on the shape. The purpose of mathematical morphology is to study the shape in the time domain, so the morphology is suitable for image processing. Corrosion, expansion, opening and closing in morphological operations are operations based on sets. The structural elements play a very important role in these operations, which adjust the geometry of the image feature transform. The image edge detection operator can be introduced by means of morphological operations. The expansion and erosion operations in mathematical morphology have a very intuitive geometric background, which can make the processed image thicker or thinner in a certain direction. The difference between the original image and the two operations can be used as a full-scale edge. Detect, ie or you can check out the edges of the image. In addition, the morphological method can also modify the edge of the obtained image by adaptive method, and gradually adjust the size of the structural element window to achieve the purpose of effectively enhancing the blurred edge and appropriately eliminating the influence of noise.

2.4 subpixel edge detection algorithm

The above edge detection algorithms are all performed at the pixel level, and sub-pixel edge detection refers to decomposing pixels near the edge to accurately determine the edge. Subpixel edge detection maps image data to a Hilbert space of nine parameters to determine edge parameters. Ghosal and Mehrotal first proposed the use of Zernike Moments (Zernike Moments ZMs) to detect sub-pixel edges. In their algorithm, an ideal step gray scale model was established for the edges. By calculating the ZMs of three different orders of the image, The four parameters of the ideal step gray scale model are mapped into three ZMs, and then the parameters of the line where the edge is located are calculated by the three ZMs, thereby determining the sub-pixel level coordinates of the edge. Li Jinquan [12] conducted a more in-depth study on the ZMs algorithm, and pointed out its inadequacies and proposed a corresponding improved algorithm, which was applied to the weld seam identification. The detected edges have high precision, self-refinement edges and Strong anti-interference and other advantages.

3. Conclusion

Most welds do not change directionally, they are continuous straight lines or curves. In the local small range can be seen as two parallel lines. Therefore, the weld seam can be detected by finding a straight line when the welding environment is identified. In these existing algorithms, the gradient operator can detect the edge of the weld while predicting the direction, so that the real-time image processing can also predict the direction of the weld advancement, which is more suitable for the identification of the welding environment. However, the differential operator has poor anti-interference. For complex welding environments, it cannot be directly applied, when it is improved and combined with other algorithms. Wavelet multi-scale and morphological edge detection algorithms are one of the research hotspots in this field. Their characteristics are suitable for complex welding environment identification and should be studied in depth. Some sub-pixel detection algorithms can obtain more accurate detection results, which is one of the efforts to improve image processing accuracy and welding results. There are countless edge detection methods, and they all have their specific scope of application. When selecting or developing new algorithms, we must consider the characteristics of the solder itself.

Glass Door made up of panes or panels of glass. Wooden frames or partitions separate each pane. Glass panes are generally rectangular in shape and make up the majority of the surface of the door.

It is typically used as a style element.

A frequently way to bring more natural light into home or to divide rooms and set them apart while still maintaining an open feel of the whole space.

For aesthetic purposes, create dramatic effect.

Connecting the interior and the exterior living space.

Glass Door

Glass Door,Kitchen Glass Door,Melamine Glass Door,Modern Glass Door

Zhejiang JiHengKang (JHK) Door Industry Co.,LTD , http://www.zjjhk.com