This article provides a comprehensive survey of Deep Active Learning (DeepAL), a field that combines the strengths of deep learning (DL) and active learning (AL) to maximize model performance while minimizing the cost of labeling samples. The authors highlight the challenges and opportunities in integrating AL and DL, particularly in handling high-dimensional data and processing pipelines. They review existing DeepAL methods, categorizing them into query strategies, model training methods, and generalization techniques. Key query strategies include batch mode DeepAL (BMDAL), uncertainty-based and hybrid query strategies, deep Bayesian active learning (DBAL), density-based methods, and automated design approaches. The article also discusses the impact of dataset attributes on DeepAL performance and suggests future research directions. Overall, the survey aims to fill the gap in a unified classification framework for DeepAL-related works, providing a systematic overview and analysis of the field.This article provides a comprehensive survey of Deep Active Learning (DeepAL), a field that combines the strengths of deep learning (DL) and active learning (AL) to maximize model performance while minimizing the cost of labeling samples. The authors highlight the challenges and opportunities in integrating AL and DL, particularly in handling high-dimensional data and processing pipelines. They review existing DeepAL methods, categorizing them into query strategies, model training methods, and generalization techniques. Key query strategies include batch mode DeepAL (BMDAL), uncertainty-based and hybrid query strategies, deep Bayesian active learning (DBAL), density-based methods, and automated design approaches. The article also discusses the impact of dataset attributes on DeepAL performance and suggests future research directions. Overall, the survey aims to fill the gap in a unified classification framework for DeepAL-related works, providing a systematic overview and analysis of the field.