The real-time counting of stacked plate is extremely significant for manufacturing. However, Stacked plate are still counted by hand today, which takes long time and has poor accuracy. Hence, the paper proposes a stacked plate counting instrument based on embedded platform with a lightweight object detection model. The instrument can detect in real time the number of stacked plate at production and logistics site, which deploys the improved lightweight Faster R-CNN network to the Industrial Personal Computer(IPC).In order to improve the efficiency of feature extraction and alleviate the difficulty of small object detection, the traditional Faster R-CNN network algorithm by using lightweight network MobileNetv2 to integrate the Efficient channel attention (ECA) as the backbone network, using Spatial attention (SA) and inverted residual structure module to reconstruct the FPN structure, proposing an HIOU_Loc algorithm based on on Height intersection over union (HIOU) to remove redundant prediction boxes. The plate counting experiment on a IPC equipped with N4100 CPU. The results show that the accuracy of the plate counting algorithm proposed in this paper reaches 98.51%, and it only takes 0.31s to detect a high-resolution plate image. A quantitative calibration module is designed for the instrument in order to improve the accuracy. The instrument can reach 100% accuracy in counting stacked plate after the manual calibration module, which meets the requirements of stacked plate real-time counting in practical scenarios.