Diagnostic plots, introduced by K. S. Chan, are widely used to determine excessive water production mechanisms. In this paper, we introduce a computer vision model that is capable of segmenting and identifying multiple Chan signatures per plot, for the sake of surveillance and early screening, given that wells could exhibit diverse mechanisms throughout their life cycle. As deep learning (DL) demands a vast amount of information, we start our workflow by building a data set of 10,000 publicly available oil wells that have experienced varying water production mechanisms. Next, we perform preprocessing and remove anomalies from production data, which could be deceptive in analysis. Then, we visualize Chan plots as images and annotate them. Thereafter, we split data set, carry out augmentation, and put data together to be used as input for a convolutional neural network (CNN) layer. Eventually, data are trained utilizing you only look once (YOLO)—a one-stage object detector, hyperparameters are tuned, and model performance is evaluated using mean average precision (mAP). The collected data from fields in Alaska and North Dakota represent oil wells that have been producing for decades. When working with wells that possess noisy production data, we recognized challenge, bias, and tedium in human interpretation of Chan plots. Subsequently, we observed the inevitability of cleaning well production data before constructing the plots and thoroughly revealed its effect on enhancing model potentiality to get a fair score. In addition, we concluded that following a systematic approach of active learning, a technique that allows user to analyze mistakes of algorithm predictions and label data accordingly, accomplished a significant boost in model performance, especially with underrepresented classes. The proposed CNN model, which uses automatic feature extraction and expresses data in detail, is presumed to be robust as it successfully predicted multiple mechanisms of excessive water production, with confidence scores higher than 80%, in wells that exhibit different production conditions such as horizontal trajectories, artificial lift, waterflooding, stimulation, and other well intervention events. In this work, we introduce a novel computer vision model, which combines image processing and DL techniques to identify multiple water production signatures that a well can undergo and eliminate subjectivity of human interpretation. This approach has the potential to be effective, as a part of workflow automation, in expeditious surveillance of large oil fields. Source code is available on GitHub for public use.