Data compression method, data compression device, solid-state imaging device, data decompression device and electronic system

1. A data compression method, the method comprising:

receiving a raw binary number comprising a first predefined number n1 of bits, each bit representing either zero or 2 (n−1) with n representing a position of the bit in the binary number and 1≤n≤n1;

identifying, in the raw binary number, a position of a highest-weighted bit unequal zero;

generating, by using a first electronic circuit, binary position information by encoding the identified position;

compiling an encoded binary number, wherein the encoded binary number is based on the binary position information and on a partial binary number being a portion of the raw binary number directly following the highest-weighted bit unequal zero, and wherein the encoded binary number comprises a second predefined number m1 of bits, wherein the second predefined number m1 of bits is smaller than the first predefined number n1 of bits.

2. The data compression method according to claim 1 ,

wherein the encoded binary number is compiled to comprise the binary position information and a copy of the portion of the raw binary number directly following the highest-weighted bit unequal zero.

3. The data compression method according to claim 1 ,

wherein, if a difference between the second number m1 of bits and a number k of bits representing the binary position information is smaller than a number of bits in the raw binary number from and including a least significant bit to and excluding the highest-weighted bit unequal zero, a rounded binary number is obtained by increasing by one an auxiliary binary number obtained by expanding the partial binary number by a next less significant bit and then truncating, by one bit, the increased auxiliary binary number, and

wherein the encoded binary number contains the binary position information and a copy of the truncated auxiliary binary number.

4. The data compression method according to claim 3 ,

wherein a variable number k of bits used for the binary position information is selected depending on the position of the highest-weighted bit unequal zero in the raw binary number.

5. The data compression method according to claim 1 ,

wherein, if in the raw binary number the position x of the highest-weighted bit unequal zero is lower than m0, with m0<m1, for each position m with 1≤m≤m0, the m-th bit in the encoded binary number is set equal to a corresponding m-th bit of the raw binary number.

6. The data compression method according to claim 1 ,

wherein the binary position information comprises a first number k1 of bits if in the raw binary number the highest-weighted bit unequal zero is the most significant bit of the raw binary number,

wherein the binary position information comprises a second number k2 of bits if in the raw binary number all but the least significant bit are equal zero, and

wherein the binary position information comprises a number of bits lower than the first number k1 and lower than the second number k2 if in the raw binary number a position of the highest-weighted bit is in a predefined range between the least significant bit and the most significant bit.

7. The data compression method according to claim 6 ,

wherein for at least one position p3 in a range from p1 to p2 the binary position information comprises a minimum number k3 of bits, and

wherein if in a first raw binary number the highest-weighted bit unequal zero is between p3 and the most significant bit and if in a subsequently processed second raw binary number the highest-weighted bit unequal zero is between p3 and the most significant bit, the binary position information encodes a differential information descriptive for a distance in bit positions between the highest-weighted bit unequal zero in the first raw binary number and the highest-weighted bit unequal zero in the second raw binary number.

8. The data compression method according to claim 1 ,

wherein the binary position information comprises a first number k1 of bits if in the raw binary number the highest-weighted bit unequal zero is any of n3 most significant bits of the raw binary number, wherein INT((m1)/2)≤n3≤((m1)−2)).

9. The data compression method according to claim 1 ,

wherein a median of a length histogram of a neuro-binary coding tree representing the decoding is smaller than the binary logarithm of the integer value of half of the second number of bits m1.

10. A data decompression method, the method comprising:

receiving an encoded binary number comprising a second predefined number m1 of bits, wherein the encoded binary number comprises a binary value and binary position information of variable length;

identifying, in the encoded binary number, the binary position information and the binary value;

obtaining, by using a second electronic circuit, a target position from the binary position information;

compiling a decoded binary number comprising a first predefined number n1 of bits, wherein the decoded binary number comprises a highest-weighted bit unequal to zero at the target position and the binary value at bit positions following the target position.

11. A data transmission method, the method comprising:

encoding, in a first electronic device, a raw binary number by using the data compression method according to claim 1 to obtain an encoded binary number;

transmitting the encoded binary number from the first electronic device to a second electronic device; and

decoding, in the second electronic device, the encoded binary number by using the data decompression method according to claim 10 to obtain a decoded binary number, wherein the decoded binary number is identical to the raw binary number for low values and approximates the raw binary number for high values.

12. A data compression device, comprising:

means for receiving a raw binary number comprising a first predefined number n1 of bits, each bit representing either zero or 2 (n−1) with n representing a position of the bit in the binary number and 1≤n≤n1;

means for identifying, in the raw binary number, a position of a highest-weighted bit unequal zero;

a first electronic circuit configured to encode the identified position to obtain binary position information;

means for compiling an encoded binary number, wherein the encoded binary is based on the binary position information and on a partial binary number being a portion of the raw binary number directly following the highest-weighted bit unequal zero, and wherein the encoded binary number comprises a second predefined number of bits m1, wherein the second predefined number m1 of bits is smaller than the first predefined number n1 of bits.

13. A solid-state imaging device, comprising:

a pixel array unit comprising a plurality of pixel circuits arranged in matrix form, the pixel circuits being configured to perform photoelectric conversion; and

an analog-to-digital converter configured to convert an electric signal corresponding to a charge obtained by the photoelectric conversion in the pixel circuit into digital pixel data; and

a data compression device according to claim 12 , wherein the data compression device is configured to use the digital pixel data as the raw binary number.

14. The solid-state imaging device according to claim 13 , further comprising:

a customary compression unit configured to encode the digital pixel data using a piece-wise linear encoding scheme.

15. A data decompression device, comprising:

means for receiving an encoded binary number comprising a second predefined number m1 of bits, wherein the encoded binary number comprises a binary value and binary position information of variable length;

means for identifying, in the encoded binary number, the binary position information and the binary value;

a second electronic circuit adapted for obtaining, from the binary position information, a target position;

means for compiling a decoded binary number comprising a first predefined number n1 of bits, wherein the decoded binary number comprises a highest-weighted bit unequal to zero at the target position and the binary value at bit positions following the target position.