{"id":25223,"date":"2024-03-14T11:47:06","date_gmt":"2024-03-14T11:47:06","guid":{"rendered":"https:\/\/difmaq.com\/?p=25223"},"modified":"2024-03-25T12:49:58","modified_gmt":"2024-03-25T12:49:58","slug":"success-case-bottle-orientation-using-machine-vision","status":"publish","type":"post","link":"https:\/\/difmaq.com\/en\/success-case-bottle-orientation-using-machine-vision\/","title":{"rendered":"Success Case: Bottle orientation using machine vision"},"content":{"rendered":"
\n
<\/div>\n\n\n\n

DIFMAQ ROURE will present the BEST CASE, carried out in Bodegas Fundador, at the next edition of HISPACK 2024 <\/h2>\n\n\n\n
<\/div>\n\n\n\n

Client<\/code><\/strong>: Bodegas Fundador, <\/a> is one of the oldest wineries in Jerez, founded in 1730. It produces three of the world’s most internationally renowned brands.<\/p>\n\n\n\n

Issue:<\/code><\/strong> In its production process, the winery had a rejection rate of over 65 bottles per thousand due to poor label quality, resulting in the need for rework at the end of each workday.<\/p>\n\n\n\n

The previous guidance system developed by an external company could only take three photographs per container.<\/p>\n\n\n\n

<\/div>\n\n\n\n

Solution:<\/code><\/strong> For this issue, DQVisio was proposed<\/a>. Our bottle orientation system detects the exact position upon entry by referencing the custom decoration and engravings on the glass, sending the result to the machine control and servo-driven bases, adjusting the bottles to the correct position.<\/p>\n\n\n\n

The outcome of this entire process enables us to place the label precisely in the designated area. It was also crucial in this case that the system could be installed on previous-generation labelling machines equipped with servo-driven bottle platforms. The new DQVisio system<\/a> allows for a complete scanning of each container using artificial vision cameras. Both the algorithm and the software have been developed entirely by DIFMAQ. <\/p>\n\n\n\n

Therefore, the improvement is not so much due to the inherent technology of the cameras, but rather the algorithm that manages their operation and the deployment of necessary actions based on the collected information. With only 3 photographs per container, there were detection gaps resulting in a high number of defective bottles. However, with our procedure, we have been able to increase the system’s precision, thus drastically reducing the rejection rate.<\/p>\n\n\n\n

Result:<\/code><\/strong> Currently, the rejection rate in the production process has decreased to 0.5%, significantly reducing the level of rejection and rework of incorrect bottles.<\/p>\n\n\n\n

<\/div>\n\n\n\n
\n