A novel vision-based mold monitoring system in an environment of intense vibration

Abstract

Mold monitoring has been more and more widely used in the modern manufacturing industry, especially when based on machine vision, but these systems cannot meet the detection speed and accuracy requirements for mold monitoring because they must operate in environments that exhibit intense vibration during production. To ensure that the system runs accurately and efficiently, we propose a new descriptor that combines the geometric relationship-based global context feature and the local scale-invariant feature transform for the image registration step of the mold monitoring system. The experimental results of four types of molds showed that the detection accuracy of the mold monitoring system is improved in the environment with intense vibration.

Aplicación del internet industrial de las cosas(ioT)en líneas de manufactura por proceso de moldeo por inyección de plástico.

“…Dado que esto es foco de interés, se han planteado a lo largo del tiempo diferentes soluciones para controlar estas variables. Resumiendo los resultados de estas implementaciones podemos ver que en (Zhang, S., Dubay, R., & Charest, M., 2014), (Tan, Y. S., Ng, Y. T., & Low, J. S. C., 2017), (Vrabič, R., Kozjek, D., & Butala, P., 2017), (Hu, F., He, Z., Zhao, X., & Zhang, S., 2017), (Siller, H. R., Romero, D., Rabelo, R. J., & Vazquez, E., 2018), (Ogorodnyk, O., & Martinsen, K., 2018), (Lucchetta, G., Masato, D., & Sorgato, M., 2018), (Ageyeva, T., Horváth, S., & Kovács, J. G., 2019), (Park, H. S., Phuong, D. X., & Kumar, S., 2019, (Lee, H., Liau, Y., & Ryu, K., 2018) y (Charest, M., Finn, R., & Dubay, R., 2018) se utiliza tecnología de sensores de temperatura y presiones en las cavidades de los moldes; cada una de las implementaciones ha tomado diferentes medidas en acondicionar las herramientas y las máquinas para poder recopilar los datos, mientras que en los casos donde se ha realizado integración de tecnología, se ha buscado que esta sea la adecuada para obtener los resultados deseados, sobre todo en (Ageyeva, T., Horváth, S., & Kovács, J. G., 2019) donde se proponen diferentes maneras de obtener los datos del comportamiento de plástico por medio de tecnologías diversas.…”

A novel solder joint diagnosis algorithm using differential geometry approach in high-density flexible integrated circuit substrates

“…Defect diagnosis for various solder joints on flexible integrated circuit substrates (FICS) is one of the critical problems in the quality control of surface-mounted technology (SMT). The performance of the diagnosis algorithms fairly impact the efficiency and yield in industrial applications [1,2]. With the continuous development of integrated circuits (ICs), the trend of miniaturization and denseness has become increasingly obvious, which makes it more difficult to extract surface features of solder joints based on acquired image [3].…”

Anomaly detection in periodic motion scenes based on multi-scale feature Gaussian weighting analysis

“…Wang et al [2] compared the foreground of the blank mold before the mold was closed with that of the empty mold and calculated the qualified rate of the region of interest (ROI) detection region of the pixel to check for foreign bodies in the mold before closing. Hu et al [12] considered that large vibrations during the mold monitoring process result in a high false alarm rate, so they proposed a method for detecting abnormalities of the injection mold or defects of the product. Butt et al [13] proposed an automatic positioning and monitoring system for sand casting molds based on machine vision.…”