361 lines
12 KiB
JavaScript
361 lines
12 KiB
JavaScript
import * as Blockly from 'blockly/core';
|
||
|
||
const AI_HUE = "#55839A";
|
||
|
||
export const tuple_anchor = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField(new Blockly.FieldTextInput('anchor'), 'VAR')
|
||
.appendField('锚点参数= (')
|
||
.appendField(new Blockly.FieldTextInput('1.889, 2.5245, 2.9465, 3.94056, 3.99987, 5.3658, 5.155437, 6.92275, 6.718375, 9.01025'), 'TEXT')
|
||
.appendField(')');
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("锚点参数");
|
||
}
|
||
};
|
||
|
||
export const tuple_calss = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField(new Blockly.FieldTextInput('calss'), 'VAR')
|
||
.appendField('物品名称= [')
|
||
.appendField(new Blockly.FieldTextInput("'name1', 'name2', 'name3', 'name4'"), 'TEXT')
|
||
.appendField(']');
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("将要识别的物品名称");
|
||
}
|
||
};
|
||
|
||
export const KPU_load = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendValueInput('SUB')
|
||
.appendField("")
|
||
.setCheck("var");
|
||
this.appendValueInput('path')
|
||
.appendField("模型加载")
|
||
.setCheck(Number);
|
||
this.setInputsInline(true);
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("从flash系统中加载模型");
|
||
}
|
||
};
|
||
|
||
export const KPU_load1 = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendValueInput('SUB')
|
||
.appendField("")
|
||
.setCheck("var");
|
||
this.appendValueInput('path')
|
||
.appendField("模型路径")
|
||
.setCheck(String);
|
||
this.setInputsInline(true);
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("从文件系统中加载模型");
|
||
}
|
||
};
|
||
|
||
export const KPU_init_yolo2 = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("yolo2")
|
||
.appendField("初始化");
|
||
this.appendValueInput('SUB')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("网络模型")
|
||
.setCheck("var");
|
||
this.appendValueInput('threshold')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("概率阈值")
|
||
.setCheck(Number);
|
||
this.appendValueInput('nms_value')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("box_iou门限")
|
||
.setCheck(Number);
|
||
this.appendValueInput('anchor_num')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("锚点数")
|
||
.setCheck(Number);
|
||
this.appendValueInput('anchor')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("锚点参数");
|
||
//this.setInputsInline(true);
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("初始化yolo2网络");
|
||
}
|
||
};
|
||
|
||
export const KPU_run_yolo2 = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("yolo2")
|
||
.appendField("运行网络");
|
||
this.appendValueInput('SUB')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("模型")
|
||
.setCheck("var");
|
||
this.appendValueInput('VAR')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("图像");
|
||
this.setOutput(true);
|
||
this.setInputsInline(true);
|
||
this.setTooltip("运行yolo2网络");
|
||
}
|
||
};
|
||
|
||
export const KPU_forward = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("yolo2")
|
||
.appendField("前向运算");
|
||
this.appendValueInput('SUB')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("模型")
|
||
.setCheck("var");
|
||
this.appendValueInput('VAR')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("图像");
|
||
this.setOutput(true);
|
||
this.setInputsInline(true);
|
||
this.setTooltip("运行网络前向运算");
|
||
}
|
||
};
|
||
|
||
|
||
export const KPU_analysis = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput()
|
||
.appendField("yolo2")
|
||
.appendField("模型解析");
|
||
this.appendValueInput('VAR')
|
||
.appendField("对象")
|
||
.setCheck("var");
|
||
this.appendDummyInput()
|
||
.appendField("获取")
|
||
.appendField(new Blockly.FieldDropdown([
|
||
["坐标-x", "x"],
|
||
["坐标-y", "y"],
|
||
["标识号", "classid"],
|
||
["置信度", "value"]
|
||
]), "key");
|
||
this.setOutput(true);
|
||
//this.setInputsInline(true);
|
||
this.setTooltip("对于模型解析，获取模型识别结果的目标坐标、标识好、置信度");
|
||
}
|
||
};
|
||
|
||
export const aionenet_nic_init = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_OneNET")
|
||
.appendField("连接WiFi");
|
||
this.appendValueInput('account')
|
||
.appendField("名称")
|
||
.setCheck(String);
|
||
this.appendValueInput('password')
|
||
.appendField("密码")
|
||
.setCheck(String);
|
||
this.setInputsInline(true);
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("AI-Onenet平台 连接WiFi");
|
||
}
|
||
};
|
||
|
||
export const aionenet_token = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_OneNET")
|
||
.appendField("获鉴权码");
|
||
this.appendValueInput('account')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("账号")
|
||
.setCheck(String);
|
||
this.appendValueInput('password')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("密码")
|
||
.setCheck(String);
|
||
this.setOutput(true);
|
||
this.setInputsInline(true);
|
||
this.setTooltip("AI-Onenet平台 需要注册平台才能使用账号获取用户鉴权码，鉴权码一般24小时有效");
|
||
}
|
||
};
|
||
|
||
export const aionenet_API = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_OneNET")
|
||
.appendField("调取API");
|
||
this.appendValueInput('VAR')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("图像");
|
||
this.appendDummyInput()
|
||
.appendField("识别")
|
||
.appendField(new Blockly.FieldDropdown([
|
||
["人脸检测", "FACE_RECO"],
|
||
["人脸分析", "FACE_ATTRIBUTE"],
|
||
["人体检测", "BODY_RECO"],
|
||
["图像抄表", "AMMETER_READ"],
|
||
["内容测评", "IDENTIFY_PORN"],
|
||
["车牌信息", "NUMBER_PLATE_RECOGNITION"],
|
||
["宠物种类", "CAT_DOG_DETECTION"],
|
||
["火灾检测", "FIRE_DETECTION"]
|
||
]), "api");
|
||
this.appendValueInput('token')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("鉴权码")
|
||
.setCheck(String);
|
||
this.setOutput(true);
|
||
this.setInputsInline(true);
|
||
this.setTooltip("AI-Onenet平台 调用平台API，返回列表识别结果参数");
|
||
}
|
||
};
|
||
|
||
export const ailocal_training = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_Local")
|
||
.appendField("模型训练");
|
||
this.appendValueInput('calss')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("物品");
|
||
this.appendValueInput('sample')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("训练量")
|
||
.setCheck(Number);
|
||
this.appendValueInput('save')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("保存")
|
||
.setCheck(String);
|
||
this.setInputsInline(true);
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("AI-Local本地模型训练 需要识别的物品名称、每个物品训练数量、保存的名称");
|
||
}
|
||
};
|
||
|
||
export const ailocal_loading = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_Local")
|
||
.appendField("模型加载");
|
||
this.appendValueInput('path')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("路径")
|
||
.setCheck(String);
|
||
this.setInputsInline(true);
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("AI-Local 加载已经训练好的本地模型");
|
||
}
|
||
};
|
||
|
||
|
||
export const ailocal_predict = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_Local")
|
||
.appendField("运行模型");
|
||
this.appendValueInput('calss')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("物品");
|
||
this.appendValueInput('VAR')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("图像");
|
||
this.setOutput(true);
|
||
this.setInputsInline(true);
|
||
this.setTooltip("AI-Local 采集图像运行模型将返回识别的物品名、置信度");
|
||
}
|
||
};
|
||
|
||
//---开始------------新增---20210302---------------------------------------------------
|
||
|
||
export const ai_face_init = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_Face")
|
||
.appendField("初始化 加载");
|
||
this.appendValueInput('FD')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("模型FD:")
|
||
.setCheck(String);
|
||
this.appendValueInput('LD')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("模型LD:")
|
||
.setCheck(String);
|
||
this.appendValueInput('FE')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("模型FE:")
|
||
.setCheck(String);
|
||
//this.setInputsInline(true);
|
||
this.setPreviousStatement(true);
|
||
this.setNextStatement(true);
|
||
this.setTooltip("人脸分辨，初始化");
|
||
}
|
||
};
|
||
|
||
export const ai_face_train = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput("")
|
||
.appendField("AI_Face")
|
||
.appendField("运行识别");
|
||
this.appendValueInput('names')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("人名");
|
||
this.appendValueInput('VAR')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("图象");
|
||
this.appendValueInput('threshold')
|
||
.setAlign(Blockly.inputs.Align.RIGHT)
|
||
.appendField("阈值")
|
||
.setCheck(Number);
|
||
this.setOutput(true);
|
||
this.setInputsInline(true);
|
||
this.setTooltip("人脸分辨，识别到人脸返回True，无人脸返回False");
|
||
}
|
||
};
|
||
|
||
export const ai_face_info = {
|
||
init: function () {
|
||
this.setColour(AI_HUE);
|
||
this.appendDummyInput()
|
||
.appendField("AI_Face")
|
||
.appendField("识别解析");
|
||
this.appendDummyInput()
|
||
.appendField("获取")
|
||
.appendField(new Blockly.FieldDropdown([
|
||
["识别人名", "info_name"],
|
||
["置信度 %", "info_score"],
|
||
["脸部坐标", "info_face"],
|
||
["三官坐标", "info_organs"]
|
||
]), "key");
|
||
this.setOutput(true);
|
||
this.setInputsInline(true);
|
||
this.setTooltip("人脸分辨,识别到人物名称，置信度，脸部坐标，三官(眼睛x2、鼻子、嘴巴*2)坐标");
|
||
}
|
||
};
|
||
|
||
//---开始------------新增---20210302---------------------------------------------------
|
||
|
||
|