This study presents a novel intelligent sensor network for object detection, classification, and recognition. Wireless sensors are utilized as the first layer to identify the coordinates of unauthorized intruders. Cameras are employed to capture image features for object classification and recognition. To reduce processing time, a novel hierarchical image extraction scheme is developed. Global object features, such as size and motion, are acquired for classification into a number of classes. When an intruder is potentially dangerous, the cameras obtain detailed image data for object recognition. Experimental results demonstrate that the proposed system detects unauthorized intrusions efficiently, classifies and recognizes intruders successfully. This application is of good use for counsumer's residence. The guard system can let them have a safer life.