| 摘要 |
The present invention relates to a method and an arrangement for suppressing noise in digital signal sequences, and to a corresponding computer program and a corresponding computer-readable storage medium, which can be used in particular to smooth and/or improve the signal-to-noise ratio in digital signal processing and in digital image processing. To this end, there is proposed a method for suppressing noise in digital signal sequences, wherein data values W0 of the signal sequence are transformed by evaluating data values Wi from an area around data values W0 to be transformed, characterized in that the number of data values Wi to be evaluated for the transformation of a data value W0 is determined individually for at least some of the data values W0 to be transformed. |
| 主权项 |
1. An arrangement comprising
at least one chip or processor which is designed in such a way that a method for suppressing noise in digital signal sequences can be carried out, wherein data values W0 of the signal sequence are transformed by evaluating data values Wi from an area around data value W0 in the sequence, and the number of data values Wi to be evaluated for the transformation of a data value W0 is determined individually for at least some of the data values W° to be transformed; wherein the arrangement is designed in such a way that the following can be carried out:
for a data value W0 of the signal sequence which is to be transformed and belongs to a point in time t0, at most m data values Wi for previous points in time ti (i=1, . . . , m−1, m) and/or at most n data values for subsequent points in time tj (j=−n, −(n−1), . . . , −1) are evaluated, according to the following rule:
successive determination, for previous data values W1 to at most Wm, of the difference between the data value Wi to be evaluated and the data value W0 to be transformed and the taking into account of the value Wi for noise suppression if the absolute value of the difference is less than or equal to a first predefinable value, and otherwise termination of the test in respect of previous data values, and/orsuccessive determination, for subsequent data values W−1 to at most W−n, of the difference between the data value Wj to be evaluated and the data value W0 to be transformed and the taking into account of the value Wj for noise suppression if the absolute value of the difference is less than or equal to a second predefinable value, and otherwise termination of the test in respect of subsequent data values,determination of the result value for the point in time t0 using the data values to be taken into account for noise suppression. |
