Nowadays, machine learning projects have become more and more relevant to various real-world use cases. The success of complex Neural Network models depends upon many factors, as the requirement for structured and machine learning-centric project development management arises. Due to the multitude of tools available for different operational phases, responsibilities and requirements become more and more unclear. In this work, Machine Learning Operations (MLOps) technologies and tools for every part of the overall project pipeline, as well as involved roles, are examined and clearly defined. With the focus on the inter-connectivity of specific tools and comparison by well-selected requirements of MLOps, model performance, input data, and system quality metrics are briefly discussed. By identifying aspects of machine learning, which can be reused from project to project, open-source tools which help in specific parts of the pipeline, and possible combinations, an overview of support in MLOps is given. Deep learning has revolutionized the field of Image processing, and building an automated machine learning workflow for object detection is of great interest for many organizations. For this, a simple MLOps workflow for object detection with images is portrayed.

RT-DETR (Real-Time DEtection TRansformer) has recently emerged as a promising model for object detection in images, yet its performance on small objects remains limited, particularly in terms of robustness. While various approaches have been explored, developing effective solutions for reliable small object detection remains a significant challenge. This paper introduces an adapted variant of RT-DETR, specifically designed to enhance robustness in small object detection. The model was first designed on one dataset and subsequently transferred to others to validate generalization. Key contributions include replacing components of the feed-forward neural network (FFNN) within a hybrid encoder with Hebbian, randomized, and Oja-inspired layers; introducing a modified loss function; and applying multi-scale feature fusion with fuzzy attention to refine encoder representations. The proposed model is evaluated on the Al-Cast Detection X-ray dataset, which contains small components from high-pressure die-casting machines, and the PCB quality inspection dataset, which features tiny hole anomalies. The results show that the optimized model achieves an mAP of 0.513 for small objects—an improvement from the 0.389 of the baseline RT-DETR model on the Al-Cast dataset—confirming its effectiveness. In addition, this paper contributes a mini-literature review of recent RT-DETR enhancements, situating our work within current research trends and providing context for future development.

Optical systems in the real world often create image problems, such as Gaussian blur from defocus or atmospheric turbulence, and radial vignetting caused by lens shape. Standard mixed-data fine-tuning helps the task head handle these issues, but it does not actually fix them. We introduce the Iterative Correction of Optical Perturbations ICOP framework, which corrects encoder feature representations before they reach the task head using a physics-aware plug-in module. ICOP does this by modeling blur as an isotropic-Gaussian point-spread function (PSF) and uses gradient-based, self-supervised optimization (Adam) to discover feature-space corrections that steer degraded representations toward their clean-data distribution. It includes a BlurEstimator that builds a degradation descriptor using fixed Laplacian and Sobel operators, and a PromptGenerator that turns this descriptor into modulation parameters for the frozen encoder output. The framework comes in two versions based on the task: an additive correction (ICOP-Add) for image classification, and a Feature-wise Linear Modulation correction (ICOP-FiLM) for object detection. We observe a convergence between clean-task performance and blur-induced degradation across datasets, consistent with greater reliance on high-frequency features in stronger backbones; we treat this as an empirical, cross-dataset observation rather than a demonstrated causal claim (task difficulty, category structure, texture, and object scale also differ across datasets). Independently of this, ICOP-FiLM’s corrective benefit does not scale with degradation severity, revealing a more nuanced relationship between backbone quality and robustness. For classification, ICOP-Add improves distorted-condition accuracy over strong mixed fine-tuning by $+2.4$, $+9.1$, and $+10.7$ percentage points on MNIST, FashionMNIST, and CIFAR-10, respectively (McNemar’s test, $p<0.001$ on all three, 5000 paired predictions per dataset). On three object detection datasets, ICOP-FiLM improves distorted-condition mAP over a mixed-fine-tuning null hypothesis by $+0.026$, $-0.005$, and $+0.003$ mAP, respectively (all values mean over 3 seeds). Against a matched-blur-ratio control that isolates the correction module’s own contribution, ICOP-FiLM wins by a consistent margin on two of the three datasets ($+0.037$ and $+0.024$ mAP, winning in every one of 3/3 seeds on each) and loses on the third ($-0.039$ mAP, losing in 3/3 seeds); it outperforms parameter-efficient (VPT, Adapter) baselines trained on identical data on the same two datasets. This dataset-dependent pattern is discussed in detail in the main text. ICOP-FiLM adds only 82,672 parameters to a 42-million-parameter Real-Time DEtection TRansformer (RT-DETR) detector. All reported results are obtained under synthetic Gaussian blur and radial vignetting applied to clean images from the six benchmark datasets studied.