The difficulty of collecting Industrial Data Digits can vary based on factors like data quality and quantity, digit recognition complexity, and the specific industrial environment. This article utilizes a digital display panel dataset to detect and recognize seven-segment numerals, aiming to reduce industry costs through wireless data transfer. The dataset comprises normal, blurred, and flickered images. To collect real-time data, the ESP32 Camera Module is employed, using video cameras positioned in front of display panels to capture streaming video. The process involves employing the YOLO (You Only Look Once) v5 Model, a real-time object detection algorithm widely used in computer vision and deep learning. YOLO accurately and quickly identifies multiple objects of different classes in images or video streams, providing bounding box coordinates and class labels for each detected object. Preprocessing raw data is crucial for model accuracy, involving techniques such as Sharpen Image for image enhancement. The YOLO v5 Model is trained on a custom dataset and subsequently tested on unseen images. The model's final output includes bounding boxes that identify the detected objects in the test images.