optimized yolo model for accurate and real-time detection of machinery around shovels in copper mining

Shovels are among the most important equipment in open-pit mining operations, widely used for loading minerals. these heavy machines play a crucial role in operational efficiency, but due to the operator’s visibility limitations, particularly in the shovel’s blind spots, they pose significant safety risks. in these situations, operators may face challenges in detecting vehicles around the shovel, increasing the likelihood of accidents and incidents. this study proposes an enhanced version of the yolo model for the precise and rapid detection of vehicles around the shovel in copper mining environments. the proposed model, using real-time processing, is capable of detecting vehicles in four directions around the shovel and preventing collisions. to evaluate this model, real-world data collected from four cameras installed around the shovel in a copper mine under various lighting conditions, including day and night, were used. the proposed method was evaluated on a new dataset of shovels under real working conditions. the results, with an average accuracy of 94.2% and a rate of 159 fps, demonstrate a significant improvement in detection accuracy and an increase in the speed of the recognition process, meeting the requirements for accurate and real-time detection of vehicles around the shovel. the findings show that the proposed model can act as an effective collision avoidance system, preventing collisions between the shovel and surrounding vehicles, which directly enhances the safety of the work environment and personnel. furthermore, this system can help reduce accidents and injuries caused by collisions between shovels and surrounding vehicles, thereby improving the overall productivity of mining operations in copper mines.

optimized yolo model for accurate and real-time detection of machinery around shovels in copper mining

shovels are among the most important equipment in open-pit mining operations, widely used for loading minerals. these heavy machines play a crucial role in operational efficiency, but due to the operator’s visibility limitations, particularly in the shovel’s blind spots, they pose significant safety risks. in these situations, operators may face challenges in detecting vehicles around the shovel, increasing the likelihood of accidents and incidents. this study proposes an enhanced version of the yolo model for the precise and rapid detection of vehicles around the shovel in copper mining environments. the proposed model, using real-time processing, is capable of detecting vehicles in four directions around the shovel and preventing collisions. to evaluate this model, real-world data collected from four cameras installed around the shovel in a copper mine under various lighting conditions, including day and night, were used. the proposed method was evaluated on a new dataset of shovels under real working conditions. the results, with an average accuracy of 94.2% and a rate of 159 fps, demonstrate a significant improvement in detection accuracy and an increase in the speed of the recognition process, meeting the requirements for accurate and real-time detection of vehicles around the shovel. the findings show that the proposed model can act as an effective collision avoidance system, preventing collisions between the shovel and surrounding vehicles, which directly enhances the safety of the work environment and personnel. furthermore, this system can help reduce accidents and injuries caused by collisions between shovels and surrounding vehicles, thereby improving the overall productivity of mining operations in copper mines.