This dataset contains automated license plate recognition detections of vehicles crossing a U.S./Mexico land border checkpoint, providing a continuous ground-truth record of vehicle movements. Each record includes plate text, issuing region, vehicle type, make, model, color, timestamp, camera source, lane identifier, and an associated raw image frame for visual verification and computer vision workflows. Plate region is explicitly flagged as U.S. or Mexico, enabling direct analyses of cross-border composition and directional flows. Multi-camera coverage yields high capture rates across lanes and times of day. The data supports temporal volume analysis, origin-destination composition, vehicle-type segmentation, anomaly detection, and training or benchmarking of vision models using real-world imagery.

Key attributes include:

• Timestamped vehicle detections with license plate text and issuing region
• Vehicle profile: type, make, model, and color
• Camera source and lane identifiers for lane-level and checkpoint analysis
• Raw image frames per record for verification and computer vision use
• Continuous coverage enabling peak detection, composition analysis, and trend monitoring

Border authorities, transportation planners, logistics operators, and CV developers gain an accurate, vehicle-level view of cross-border traffic composition and timing to optimize operations, plan capacity, and train or validate vision models.