Text Detection is available in intuVision Text module.
Text Detection can be used alone to detect and recognize any text in the scene, or in conjunction with another intuVision VA event to get the text on objects triggering alarms for those events. The detected text is logged as any other intuVision VA detections, including date, time, camera, and snapshot for verification. Text outputs can be searched as words, letters and numbers in the intuVision Review application.
Text detection options include the sensitivity level and if just the letters or just numbers, or both should be expected in the text. Sensitivity level allows calibrating text detection to the scene. A lower sensitivity would work if the text is expected to be large and close to the camera, and a higher sensitivity if the text could be smaller, further from the camera, or might not be clear.
Compatible with a wide variety of text sizes and fonts, and with ability to detect text up to 30 degrees, text detector offers the flexibility to work in many scenes, environments, and domain areas.
Multiple zone selection allows detecting text in specific areas of the scene. In busy environments, with several texts are visible, such as trucks with many labels, vehicle IDs, and signs present, zones allow filtering the irrelevant information.
In the manufacturing domain, intuVision VA Text Detection can be used to output labels from containers, trucks, and machinery. These labels can be searched by full or partial text, date, time, and selected cameras. Text detection is an ideal solution many uses, whether a record of every package is needed, searchable alerts for each vehicle which took too long in the loading bay.
In the parking and traffic domains, intuVision VA Text Detection can be used to save the relevant text for vehicle identification including the letters and numbers on the license plate. For example, if a vehicle stopping in the fire lane is detected, the license plate can be output for easy intervention or follow up.
In the retail domain, text detection can be used for to manage loading bay usage by different vendor vehicles, or if looking to A/B test the efficacy of ad signage, the dwell event can be tied to text detection. In this way, typing in a key word from each sign would output all events recorded during the period time while a particular signage was up– automating that aspect of the test.
Optional settings are available to fine-tune the event, including:
When combined with compound events and output trigger options, Text Detection can be further refined, to be used for meeting custom video analytics requirements with our off-the-shelf product without the need for any development time. For example, Text Detection can be used in conjunction with any intuVision VA event to output only the text relevant to that event.