Advanced Object Detection with YOLOv12

I'm seeking an experienced developer to create an object detection project based on YOLOv12, enhanced by Slime Mould Algorithm and Marine Predators Optimization. The dataset consists of high-resolution drone images aimed at detecting crops, weeds, and soil conditions.

Requirements:
- Develop YOLOv12 model and enhance it using Slime Mould Algorithm and Marine Predators Optimization.
- Detect and analyze crops, weeds, and soil conditions.
- Provide comprehensive efficiency parameters including:
- Mean IoU
- Confidence score
- Localization error
- FPS
- BLE
- Learning rate
- Create detailed graphs such as:
- IoU distribution
- Detection confidence score distribution
- Localization error vs object size
- FPS vs iteration
- mAP50 across iterations
- Processing time vs object count
- BLE vs object size
- Visual output required: Efficiency parameters graphs and some real visualization images.

Ideal Skills and Experience:
- Expertise in YOLO object detection models
- Strong background in machine learning and optimization algorithms
- Proficient in Python and relevant libraries (TensorFlow, PyTorch, OpenCV)
- Experience with high-resolution image processing
- Ability to create and interpret complex performance metrics and graphs

Please provide examples of similar work done.