Outline of object recognition

The term object detection describes methods to identify a known object within an object space by means of optical, acoustic or other physical detection method. For example, the presence of an object in an image or a range is determined, but also the position and location.

Applications

An object recognition is necessary, for example in complex manufacturing processes. This can be used for both optical and inductive, capacitive or magnetic sensors and systems. Often, here the conformity of the shape of an object is determined by a default, or the correct location of the same on a conveyor belt checked.

In the abstract image processing object recognition methods are used to distinguish certain objects or patterns of other objects. For this purpose, the actual object must first be described mathematically. Frequently used methods of image processing, the edge detection, transformations, as well as size and color recognition. The more accurate the description of the object is possible and the more analyzable information is available, the more reliable works in object recognition.

Labels, which are available in red or blue versions, can be distinguished by a simple color sensor, for example, already. But should also be recognized, whether a font is applied to the red label, so usually a corresponding camera must be used.

In vehicles increasingly camera-based driver assistance systems are used to detect, for example, traffic signs or the lane automatically.

In the detection of traffic signs are searched for circular objects. A red box then indicates a prohibition sign. Alternatively, can also be compared only with patterns of known characters ( template matching ).

More complex forms of object recognition are then used when rotating or fast moving objects need to be tracked in an image. Then mathematical correlators are used, which can be adapted to the object, as must be determined, for example, in a helicopter to wegdrehenden against complex background, its position. By means of suitable image processing systems objects and object movements in the area and below the pixel resolution of images are possible.