Object detection methods based on fully convolutional neural networks (FCNN), such as single shot multiple box (SSD), real-time object detection (YOLO), and region-based fully convolutional networks (R-FCN), provides better performance than previous detection methods using hand-craft features. The FCNN generates more rich feature hierarchies for accurate object detection by establishing several convolution and pooling layers. These methods aim to detect and localizes multi-class objects in images by training classification and box regression models. In this paper, we focus on improving pedestrian detection performance by integrating a FCNN-based object detection method and a hand-craft feature-based method.