Electrophotographic process control device using a neural network to control an amount of exposure

Claim 2

An electrophotographic process control device, comprising: a neural network having an input layer and an output layer; sensors disposed within and in peripheral portions of an electrophotographic process mechanism and including a surface electrometer, a drum ammeter, a drum rotation counter, an exposure duration counter, a copy counter, a temperature sensor and a humidity sensor; wherein at a learning stage of said neural network, data from said sensors are applied to said input layer of said neural network, and a latent image gamma characteristic indicative of a relation between an amount of exposure and a potential of an image area and which is one of characteristics of a photoconductive drum to be used in an electrophotographic process is used as learning data to be given via said output layer of said neural network; and wherein at a control stage of said neural network, the data from said sensors are applied to said input layer of said neural network in the same manner as at said learning stage, and an amount of exposure is so controlled as to set up a desired potential in an image area on the basis of a latent image gamma characteristic obtained from said output layer.

Claim 3

An electrophotographic process control device, comprising: a neural network having an input layer and an output layer; sensors disposed within and in peripheral portions of an electrophotographic process mechanism and including a surface electrometer, a drum ammeter, a drum rotation counter, an exposure duration counter, a copy counter, a temperature sensor and a humidity sensor; wherein at a learning stage of said neural network, data from said sensors are applied to said input layer of said neural network, and a charge characteristic indicative of a relation between a potential of a background and a charge grid potential and which is one of characteristics of a photoconductive drum to be used in an electrophotographic process is used as learning data to be given via said output layer of said neural network; and wherein at a control stage of said neural network, the data from said sensors are applied to said input layer of said neural network in the same manner as at said learning stage, and an amount of exposure is so controlled as to set up a desired potential in an image area on the basis of a charge characteristic obtained from said output layer.

Claim 4

An electrophotographic process control device, comprising: a neural network having an input layer and an output layer; sensors disposed within and in peripheral portions of an electrophotographic process mechanism and including a surface electrometer, a drum ammeter, a drum rotation counter, an exposure duration counter, a copy counter, a temperature sensor and a humidity sensor; wherein at a learning stage of said neural network, data from said sensors are applied to said input layer of said neural network, and a charge characteristic indicative of a relation between a potential of a background and a charge grid potential and which is one of characteristics of a photoconductive drum to be used in an electrophotographic process is used as learning data to be given via said output layer of said neural network; and wherein at a control stage of said neural network, the data from said sensors are applied to said input layer of said neural network in the same manner as at said learning stage, and an amount of exposure is so controlled as to set up a desired potential in an image area on the basis of a charge characteristic obtained from said output layer.

Claim 5

An electrophotographic process control device, comprising: a neural network having an input layer and an output layer; sensors disposed within and in peripheral portions of an electrophotographic process mechanism and including a surface electrometer, a drum ammeter, a drum rotation counter, an exposure duration counter, a copy counter, a temperature sensor and a humidity sensor; wherein at a learning stage of said neural network, data from said sensors are applied to said input layer of said neural network, and a latent image gamma characteristic indicative of a relation between an amount of exposure and a potential of an image area and which is one of characteristics of a photoconductive drum to be used in an electrophotographic process and a charge characteristic indicative of a relation between a potential of a background and a charge grid potential and which is another characteristic of said photoconductive drum are used as learning data to be given via said output layer of said neural network; and wherein at a control stage of said neural network, the data from said sensors are applied to said input layer of said neural network in the same manner as at said learning stage, and an amount of exposure is so controlled as to set up a desired potential in an image area on the basis of a latent image gamma characteristic and a charge characteristic obtained from said output layer.

Claim 6

An electrophotographic process control device, comprising: a neural network having an input layer and an output layer; sensors disposed within and in peripheral portions of an electrophotographic process mechanism and including a surface electrometer, a drum ammeter, a drum rotation counter, an exposure duration counter, a copy counter, a temperature sensor and a humidity sensor; wherein at a learning stage of said neural network, data from said sensors are applied to said input layer of said neural network, and a latent image gamma characteristic indicative of a relation between an amount of exposure and a potential of an image area and which is one of characteristics of a photoconductive drum to be used in an electrophotographic process and a charge characteristic indicative of a relation between a potential of a background and a charge grid potential and which is another characteristic of said photoconductive drum are used as learning data to be given via said output layer of said neural network; and wherein at a control stage of said neural network, the data from said sensors are applied to said input layer of said neural network in the same manner as at said learning stage, and an amount of exposure is so controlled as to set up a desired potential in an image area on the basis of a latent image gamma characteristic and a charge characteristic obtained from said output layer.