In today’s industrial production process, in order to keep the product quality unchanged or improve and keep the production operation in a continuous and stable state, the real-time monitoring of process variables of product quality becomes more and more important. Although the development of measurement technology makes the previously unmeasurable variables become measurable, some key quality variables are still difficult to meet the requirements of real-time measurement due to the bad measurement environment, high cost of instruments and the time lag caused by assay. Therefore, there is an urgent need for a processor to predict the reliability of complex and changeable state information, keep the reliable and useful state information, and realize the reliability measurement of state information, so as to provide people with targeted guidance and help. The information content in digital image is divided into perceptual content and semantic content. Perceptual content includes color, shape, texture, frequency, material and time-domain change; semantic content includes target, event and its relationship. Text is a special target which contains rich semantic information. Text analysis is the key clue to describe and understand the content of scene. Therefore, this paper uses visual imaging technology and text analysis to build an efficient soft computing model based on Bayesian network.

在当今的工业生产过程中，为了保持产品质量不变或改善并保持生产操作连续稳定，对产品质量过程变量的实时监控变得越来越重要。尽管测量技术的发展使以前无法测量的变量变得可以测量，但是由于恶劣的测量环境，高昂的仪器成本以及检测所带来的时间滞后，一些关键的质量变量仍然难以满足实时测量的要求。因此，迫切需要一种处理器，以预测复杂多变的状态信息的可靠性，保持可靠且有用的状态信息，并实现状态信息的可靠性测量，从而为人们提供有针对性的指导和帮助。数字图像中的信息内容分为感知内容和语义内容。感知内容包括颜色，形状，纹理，频率，材料和时域变化；语义内容包括目标，事件及其关系。文本是包含丰富语义信息的特殊目标。文本分析是描述和理解场景内容的关键线索。因此，本文利用视觉成像技术和文本分析技术，建立了基于贝叶斯网络的高效软计算模型。文本是包含丰富语义信息的特殊目标。文本分析是描述和理解场景内容的关键线索。因此，本文利用视觉成像技术和文本分析技术，建立了基于贝叶斯网络的高效软计算模型。文本是包含丰富语义信息的特殊目标。文本分析是描述和理解场景内容的关键线索。因此，本文利用视觉成像技术和文本分析技术，建立了基于贝叶斯网络的高效软计算模型。