Research Article
Automated Degradation Diagnosis in Character Recognition System Subject to Camera Vibration
Algorithm 1
Motion-blurring diagnosis procedure.
Input: an input character image | Output: motion-blurring level | Step 1. A given character image is preprocessed to a uniform image ( pixels). | Four edge structure elements are used on to acquire the edge images , , , . | Step 2. The edge density ratio feature and can be computed by (i) and (ii): | (i) | (ii) | Here, , , and are the respective gray distribution features of | , , , , namely if the bin of | gray histogram is more than defined thresholding , it would be recorded . | Step 3. The one big level after disk-blurring diagnosis can be further diagnosed as followed: | if | elseif | elseif | elseif | else | Here , , and are the thresholdings for motion-blurring diagnosis in four directions. |
|