Ash content detection of coal slime flotation tailings is a crucial aspect of flotation process intelligence. To enhance the accuracy and stability of ash content detection in flotation tailings and address the susceptibility to failure of ash prediction methods based solely on surface slurry images in complex coal preparation plants, this paper presents the design of a novel image acquisition device inspired by the hand movements of on-site operators. The core of this device is an anthropomorphic motion module driven by dual motors. Its design adheres to the fourth strength theory of material mechanics and employs maximum deflection as the criterion for verifying structural strength and rigidity. Motion optimization is achieved using quintic polynomial S-curve velocity planning to ensure mechanical integrity, operational smoothness, and impact reduction. The device integrates an image acquisition module comprising a CCD camera and a ball screw feed system to capture images of both the slurry surface and settled solids. It also incorporates a conveying pipeline, an oil film removal module, and a sample retention bucket. The oil film removal module exploits the density difference between slurry and oil film for preliminary separation, followed by residual oil film removal via airflow. An experimental platform was constructed for validation. Results demonstrate that the device can stably acquire images of tailings slurry surfaces and sediments, featuring a rational structural design and performance achieving parity with manual operation levels.