With the rapid development of computer technology, image measurement technology has been widely used because of its non-contact, high precision and other advantages. In recent years, potable consumer devices like mobile phone, VR/AR device, tablet computer etc. become more and more popular. The number of high precision small workpieces has increased exponentially. There is an urgent need for the high accuracy and high speed measurement approach for the quality control of the small workpieces. Due to the contradiction between accuracy and field of view, the existing image measuring instruments are not sufficient, and it is difficult to meet the needs of high precision and high efficiency measurement. In this paper, several aspects of high-precision image measurement system are explored, including hardware system construction, double side telecentric lens design, high-precision template matching technology, system calibration, sub-pixel feature extraction and image segmentation algorithm. According to the actual demand, a high precision detection system is integrated, a high-precision image rapid measurement system for small workpieces (size < 70mm) is designed and developed. The accuracy of 2um and speed of 7000UPH are obtained which can meet the industrial requirements.
The traditional Canny algorithm has the problem of edge loss in the process of smoothing the image and needs to set up the high and low threshold, an improved adaptive edge detection algorithm based on Canny is proposed in this paper. Firstly, the improved anisotropic diffusion filter is used to smooth the image, and the edge is protected when de-noising. Then, 4 gradient templates in horizontal direction, vertical direction, 45° direction, and 135° direction are used to calculate gradient amplitude. Finally, the threshold is adaptively determined according to the gray histogram of image. Experimental results indicate that the proposed algorithm has better anti-noise performance while detecting more edge details.
Solder ball detection in full-field ball grid array (BGA) images has a broad range of applications, such as height extraction of solder ball, inspection of substrate coplanarity, and defective detection. Existing methods usually have poor performance due to the diversity of defects, image noise, and the disturbances of background. In this paper, we propose a coarse-to-fine process for solder ball detection by combing the strength of the threshold method and the active contour method. In the coarse process, the solder ball is roughly segmented by a simple threshold method. In the fine process, the region information and shape prior are integrated into the energy function of the active contour method to better segment the solder ball. The initial shape used in the fine process can be given by the simple threshold method in the coarse process. Experiments on full-field BGA images demonstrate the robustness and accuracy of our method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.