Source detection is an important part of data processing pipeline, and it is also one of the challenges that Square Kilometre Array (SKA) and other next generation telescopes are facing with when dealing with massive data. To date, source detection algorithms have become quite mature and been applied in various data processing. Meanwhile, there are still areas such as automation that can be further improved on, and more tests are necessary for the fulfillment of the requirement of SKA data processing. The purpose of this paper is to conduct the research on source detection algorithms that are more automated and adaptive to massive data processing. Based on it, the research team has made improvements of source detection algorithm, and designed and developed a set of automated source detection software system, which is highlighted with a user-friendly interactive interface, output display function, more compatible data input and output, and improved data management. Integrating multiple functions together enables it to have good performance for automatic processing of large sky image and image sets, and the test results show that the improvements are effective. The research team will make further improvements and develop functions to meet the needs of SKA.