Abstract: This paper proposes an automated detection method based on an improved YOLOv5 model to address the subjectivity and low efficiency of traditional chloride ion titration endpoint discrimination methods in durability testing of concrete structures. This method introduces an inverted residual structure in the YOLOv5 backbone network to enhance the model's perception ability of small targets and subtle color changes, and integrates a spatial channel dual attention mechanism to dynamically focus on the feature information of key regions during the titration process. To verify the effectiveness of the model, this article also constructed a chloride ion titration dataset containing 3000 images, which covers the entire process scenarios such as before titration, during titration, after titration, and boiling. All images undergo HSV color space conversion to reduce the impact of lighting conditions on the images, and various data augmentation methods are combined to improve the model's generalization ability. After bounding box annotation, the dataset is divided into training set, validation set, and testing set in a ratio of 7:1.5:1.5. During the model training process, the stochastic gradient descent (SGD) optimization algorithm is used, and momentum terms and L2 regularization strategies are introduced to accelerate model convergence and suppress overfitting. The experimental results show that the improved model achieved excellent performance on the test set: mAP@0.5 Reaching 99.0%, the accuracy of endpoint detection is 99.5%, and the recall rate is 93.7%. Especially under complex conditions such as boiling solutions and bubble interference, the model exhibits good robustness, with an accuracy improvement of 6.8% compared to traditional methods. This study has achieved the transition from manual identification to intelligent detection of chloride ion titration endpoints, providing an efficient and accurate solution for concrete durability evaluation, as well as important reference and technical support for automated detection in industrial quality inspection and chemical analysis scenarios.