Automatic Intermodal Loading Unit Identification using Computer Vision: A Scoping Review

TL;DR

This scoping review maps CV-based approaches for Intermodal Loading Unit Identification, clarifies terminology, and traces methodological evolution from early DIP/ML methods to modern DL pipelines. It reveals a strong Asia-centric research footprint, a shift toward vehicle-mounted sensing, and wide variation in end-to-end performance due to limited public datasets and benchmarks. Key contributions include a proposed inclusive ILU Identification terminology, a synthesis of datasets and evaluation practices, and a roadmap emphasizing synthetic data, multi-view approaches, and unified end-to-end models. The findings highlight the transformative potential for terminal operations, tempered by reproducibility and benchmarking challenges that require open datasets, open-source code, and standardized evaluation protocols.

Abstract

Background: The standardisation of Intermodal Loading Units (ILUs), including containers, semi-trailers, and swap bodies, has transformed global trade, yet efficient and robust identification remains an operational bottleneck in ports and terminals. Objective: To map Computer Vision (CV) methods for ILU identification, clarify terminology, summarise the evolution of proposed approaches, and highlight research gaps, future directions and their potential effects on terminal operations. Methods: Following PRISMA-ScR, we searched Google Scholar and dblp for English-language studies with quantitative results. After dual reviewer screening, the studies were charted across methods, datasets, and evaluation metrics. Results: 63 empirical studies on CV-based solutions for the ILU identification task, published between 1990 and 2025 were reviewed. Methodological evolution of ILU identification solutions, datasets, evaluation of the proposed methods and future research directions are summarised. A shift from static (e.g. OCR-gates) to vehicle mounted camera setups, which enables precise monitoring is observed. The reported results for end-to-end accuracy range from 5% to 96%. Conclusions: We propose standardised terminology, advocate for open-access datasets, codebases and model weights to enable fair evaluation and define future work directions. The shift from static to dynamic camera settings introduces new challenges that have transformative potential for transportation and logistics. However, the lack of public benchmark datasets, open-access code, and standardised terminology hinders the advancements in this field. As for the future work, we suggest addressing the new challenges emerged from vehicle mounted cameras, exploring synthetic data generation, refining the multi-stage methods into unified end-to-end models to reduce complexity, and focusing on contextless text recognition.

Figures (5)

• Figure 1: Schematic overview of a terminal, highlighting the use of camera-equipped aerial and ground vehicles and ocr-gates for comprehensive port monitoring.
• Figure 2: Geographic distribution of articles by country (n=63). Five articles involve international collaborations, which results in 69 for country mentions rather than 63.
• Figure 3: Ranking of publication outlets for the eligible articles ($n = 63$) over the year. The years with zero eligible publications are ommited in this barchart. As this scoping review was prepared in 2025, the data for that year remain incomplete, covering only publications up to August.
• Figure 4: The number of publications and their approach categories used for ilu identification task over the years.
• Figure 5: Sample images for fixed perspective (left) verma2016automatic, ground vehicle perspective (middle) gulsoylu2025trudi, aerial perspective (right) teegen2024drone.