admin/smartDriveEE
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

文本生成图片

Browse Source
This commit is contained in:
spllzh
2025-10-18 10:48:46 +08:00
parent 8e640e03c8
commit 9b2e08593c
25 changed files with 356 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/main/java/com/rj/common/PasswordUtil.java
View File

@@ -173,6 +173,8 @@ public class PasswordUtil {

2
src/main/java/com/rj/common/ServiceManager.java
View File

@@ -157,6 +157,8 @@ public class ServiceManager {

2
src/main/java/com/rj/config/AliyunConfig.java
View File

@@ -134,6 +134,8 @@ public class AliyunConfig {

25
src/main/java/com/rj/controller/ImageModelController.java
View File

@@ -678,6 +678,31 @@ public class ImageModelController {
}
}
@PostMapping("/image-gen-byText")
@Operation(summary = "文本生成图像", description = "基于文本提示词生成图像")
public ResponseEntity<Map<String, Object>> imageGen(
@Parameter(description = "文本生成图像") @Valid @RequestBody TextModelController.ImageProcessingRequest request) {
log.info("开始文本生成图像，request: {}", request);
try {
// 调用服务层进行文本生成图像处理
Map<String, Object> processingResponse = imageModelService.generateImageByText(request);
log.info("文本生成图像处理完成，生成图片数量: {}", processingResponse.get("imageCount"));
return ResponseEntity.ok(processingResponse);
} catch (Exception e) {
log.error("文本生成图像处理失败: {}", e.getMessage(), e);
Map<String, Object> result = new HashMap<>();
result.put("success", false);
result.put("message", "文本生成图像处理失败: " + e.getMessage());
result.put("error", e.getClass().getSimpleName());
return ResponseEntity.status(HttpStatus.INTERNAL_SERVER_ERROR).body(result);
}
}
/**
* 图像处理接口，根据给的素材生成图片
*/

6
src/main/java/com/rj/controller/TextModelController.java
View File

@@ -137,6 +137,12 @@ public class TextModelController {
@NotBlank(message = "图片名称")
private String imageName;
@NotBlank(message = "提示词")
private String prompt;
@NotBlank(message = "模型")
private String model;
private ReferenceEdge referenceEdge;
private Parameters parameters;
}

2
src/main/java/com/rj/controller/sys/MenuController.java
View File

@@ -395,6 +395,8 @@ public class MenuController {

2
src/main/java/com/rj/controller/sys/RoleController.java
View File

@@ -365,6 +365,8 @@ public class RoleController {

2
src/main/java/com/rj/controller/sys/UserRoleController.java
View File

@@ -371,6 +371,8 @@ public class UserRoleController {

2
src/main/java/com/rj/dto/DifyWorkflowResponseDto.java
View File

@@ -137,6 +137,8 @@ public class DifyWorkflowResponseDto {

2
src/main/java/com/rj/mapper/CustomerProfileAnalysisMapper.java
View File

@@ -80,6 +80,8 @@ public interface CustomerProfileAnalysisMapper extends BaseMapper<CustomerProfil

16
src/main/java/com/rj/service/IImageModelService.java
View File

@@ -1,8 +1,11 @@
package com.rj.service;
import com.baomidou.mybatisplus.extension.service.IService;
import com.rj.controller.TextModelController;
import com.rj.entity.ImageModel;
import java.util.Map;
/**
* 图像模型服务接口
*
@@ -140,4 +143,17 @@ public interface IImageModelService extends IService<ImageModel> {
String prompt, String functionName);
boolean removeByUuid(String uuid);
/**
* 文本生成图像
*
* 根据文本提示词生成图像，包括：
* - 调用AI图像生成服务
* - 处理生成结果
* - 保存生成记录
*
* @param request 文本生成图像请求参数
* @return 生成结果，包含图像URL等信息
*/
Map<String, Object> generateImageByText(TextModelController.ImageProcessingRequest request);
}

2
src/main/java/com/rj/service/biz/ICustomerProfileAnalysisService.java
View File

@@ -101,6 +101,8 @@ public interface ICustomerProfileAnalysisService extends IService<CustomerProfil

2
src/main/java/com/rj/service/biz/impl/CustomerProfileAnalysisServiceImpl.java
View File

@@ -136,6 +136,8 @@ public class CustomerProfileAnalysisServiceImpl extends ServiceImpl<CustomerProf

206
src/main/java/com/rj/service/impl/ImageModelServiceImpl.java
View File

@@ -4,10 +4,18 @@ import ai.djl.util.JsonUtils;
import com.alibaba.dashscope.aigc.imagesynthesis.ImageSynthesis;
import com.alibaba.dashscope.aigc.imagesynthesis.ImageSynthesisParam;
import com.alibaba.dashscope.aigc.imagesynthesis.ImageSynthesisResult;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;
import com.baomidou.mybatisplus.core.conditions.query.LambdaQueryWrapper;
import com.baomidou.mybatisplus.extension.service.impl.ServiceImpl;
import com.rj.controller.TextModelController;
import com.rj.entity.ImageModel;
import com.rj.mapper.ImageModelMapper;
import com.rj.service.IImageModelService;
@@ -20,9 +28,7 @@ import org.springframework.web.multipart.MultipartFile;
import java.io.*;
import java.net.URL;
import java.time.LocalDateTime;
import java.util.HashMap;
import java.util.Map;
import java.util.UUID;
import java.util.*;
/**
* 图像模型服务实现类
@@ -276,7 +282,8 @@ public class ImageModelServiceImpl extends ServiceImpl<ImageModelMapper, ImageMo
} else {
log.warn("AI处理结果为空或无效");
}
imageModel.setImageType("remove_water");
// 6. 保存处理记录
boolean saveResult = this.updateById(imageModel);
if (!saveResult) {
@@ -887,4 +894,195 @@ public class ImageModelServiceImpl extends ServiceImpl<ImageModelMapper, ImageMo
}
}
/**
* 文本生成图像
*
* 根据文本提示词生成图像，包括：
* - 调用AI图像生成服务
* - 处理生成结果
* - 保存生成记录
*
* @param request 文本生成图像请求参数
* @return 生成结果，包含图像URL等信息
*/
@Override
public Map<String, Object> generateImageByText(TextModelController.ImageProcessingRequest request) {
try {
log.info("开始文本生成图像，提示词: {}，模型: {}", request.getPrompt(), request.getModel());
Map<String, Object> result = new HashMap<>();
// 1. 参数验证
if (request.getPrompt() == null || request.getPrompt().trim().isEmpty()) {
throw new IllegalArgumentException("提示词不能为空");
}
if (request.getModel() == null || request.getModel().trim().isEmpty()) {
throw new IllegalArgumentException("模型名称不能为空");
}
// 2. 调用阿里云图像生成API
String apiKey = System.getenv("DASHSCOPE_API_KEY");
if (apiKey == null || apiKey.trim().isEmpty()) {
throw new RuntimeException("未配置DASHSCOPE_API_KEY环境变量");
}
// 3. 设置API基础URL（参考测试类）
Constants.baseHttpApiUrl = "https://dashscope.aliyuncs.com/api/v1";
// 4. 构建多模态对话参数（参考测试类）
MultiModalConversation conv = new MultiModalConversation();
MultiModalMessage userMessage = MultiModalMessage.builder()
.role(Role.USER.getValue())
.content(Arrays.asList(
Collections.singletonMap("text", request.getPrompt())
)).build();
// 5. 构建参数（参考测试类）
Map<String, Object> parameters = new HashMap<>();
parameters.put("watermark", true);
parameters.put("prompt_extend", true);
parameters.put("negative_prompt", "");
parameters.put("size", "1328*1328"); // 使用阿里云允许的尺寸
MultiModalConversationParam param = MultiModalConversationParam.builder()
.apiKey(apiKey)
.model("qwen-image-plus") // 使用固定的模型名称
.messages(Collections.singletonList(userMessage))
.parameters(parameters)
.build();
// 6. 调用图像生成服务
MultiModalConversationResult conversationResult = null;
try {
log.info("开始调用阿里云多模态对话API，参数: {}", param);
conversationResult = conv.call(param);
log.info("多模态对话API调用完成，结果: {}", JsonUtils.toJson(conversationResult));
} catch (ApiException | NoApiKeyException | UploadFileException e) {
log.error("阿里云多模态对话API调用失败", e);
throw new RuntimeException("图像生成API调用失败: " + e.getMessage(), e);
}
// 7. 处理生成结果
if (conversationResult != null && conversationResult.getOutput() != null &&
conversationResult.getOutput().getChoices() != null &&
!conversationResult.getOutput().getChoices().isEmpty()) {
log.info("图像生成成功，开始处理结果");
log.info("完整API返回结果: {}", JsonUtils.toJson(conversationResult));
// 从多模态对话结果中提取图像URL
String generatedImageUrl = null;
try {
// 根据您提供的返回结果格式解析图像URL
// 返回格式：{"output": {"choices": [{"message": {"content": [{"image": "url"}]}}]}}
var choices = conversationResult.getOutput().getChoices();
if (!choices.isEmpty()) {
var message = choices.get(0).getMessage();
var content = message.getContent();
log.info("API返回的content: {}", JsonUtils.toJson(content));
// 解析content数组中的image字段
if (content != null && !content.isEmpty()) {
// content是一个List，包含多个内容项
for (Object contentItem : content) {
if (contentItem instanceof Map) {
@SuppressWarnings("unchecked")
Map<String, Object> contentMap = (Map<String, Object>) contentItem;
if (contentMap.containsKey("image")) {
generatedImageUrl = (String) contentMap.get("image");
log.info("提取到图像URL: {}", generatedImageUrl);
break;
}
}
}
}
}
if (generatedImageUrl == null || generatedImageUrl.trim().isEmpty()) {
log.warn("未能从API返回结果中提取图像URL，完整结果: {}", JsonUtils.toJson(conversationResult));
throw new RuntimeException("未能从API返回结果中提取图像URL");
}
} catch (Exception e) {
log.error("解析API返回结果失败", e);
throw new RuntimeException("解析API返回结果失败: " + e.getMessage(), e);
}
if (generatedImageUrl != null && !generatedImageUrl.trim().isEmpty()) {
try {
log.info("开始下载生成的图片: {}", generatedImageUrl);
// 下载图片到字节数组
byte[] imageBytes = downloadImageToBytes(generatedImageUrl);
log.info("图片下载成功，大小: {} bytes", imageBytes.length);
// 生成唯一文件名
String resultFileName = generateUniqueFileName("png");
// 上传到MinIO
String resultImageMinIOUrl = minIOService.uploadFile(imageBytes, resultFileName, "image/png");
log.info("图片上传到MinIO成功: {}", resultImageMinIOUrl);
// 生成7天临时访问链接
String resultImageTempUrl = minIOService.generateTempUrl(resultFileName);
log.info("生成7天临时访问链接: {}", resultImageTempUrl);
// 保存图像模型记录到数据库
ImageModel imageModel = new ImageModel();
imageModel.setImageName(request.getImageName() != null ? request.getImageName() : "文本生成图像");
imageModel.setOwnerName(request.getOwnerName());
imageModel.setOwnerPhone(request.getOwnerPhone());
imageModel.setImageType("text_to_image");
imageModel.setPrompt(request.getPrompt());
imageModel.setModelName(request.getModel());
imageModel.setResultImageUrl(resultImageMinIOUrl);
imageModel.setResultImageTempUrl(resultImageTempUrl);
imageModel.setCreateTime(LocalDateTime.now());
imageModel.setUpdateTime(LocalDateTime.now());
// 保存到数据库
boolean saveResult = this.saveImageModel(imageModel);
if (!saveResult) {
log.error("保存文本生成图像记录失败");
throw new RuntimeException("保存文本生成图像记录失败");
}
// 构建返回结果
result.put("success", true);
result.put("message", "文本生成图像处理完成");
result.put("imageCount", 1);
result.put("imageUrl", resultImageTempUrl);
result.put("imageMinIOUrl", resultImageMinIOUrl);
result.put("prompt", request.getPrompt());
result.put("model", request.getModel());
result.put("requestId", imageModel.getUuid());
result.put("timestamp", LocalDateTime.now());
log.info("文本生成图像处理完成，图像已保存到数据库，ID: {}", imageModel.getUuid());
} catch (Exception e) {
log.error("处理生成的图片失败", e);
throw new RuntimeException("处理生成的图片失败: " + e.getMessage(), e);
}
} else {
log.warn("生成的图片URL为空");
throw new RuntimeException("生成的图片URL为空");
}
} else {
log.warn("图像生成结果为空");
throw new RuntimeException("图像生成结果为空");
}
return result;
} catch (Exception e) {
log.error("文本生成图像异常", e);
throw new RuntimeException("文本生成图像失败: " + e.getMessage(), e);
}
}
}

2
src/main/java/com/rj/service/sys/impl/MenuServiceImpl.java
View File

@@ -108,6 +108,8 @@ public class MenuServiceImpl extends ServiceImpl<MenuMapper, Menu> implements IM

2
src/main/java/com/rj/service/sys/impl/RoleServiceImpl.java
View File

@@ -108,6 +108,8 @@ public class RoleServiceImpl extends ServiceImpl<RoleMapper, Role> implements IR

2
src/main/java/com/rj/service/sys/impl/UserRoleServiceImpl.java
View File

@@ -108,6 +108,8 @@ public class UserRoleServiceImpl extends ServiceImpl<UserRoleMapper, UserRole> i

2
src/main/resources/application-audio.yml
View File

@@ -117,6 +117,8 @@ spring:

2
src/main/resources/mapper/AudioManagementStatisticsMapper.xml
View File

@@ -94,6 +94,8 @@

2
src/main/resources/static/tts-demo.html
View File

@@ -293,6 +293,8 @@

67
src/test/java/com/rj/image/QwenImage.java Normal file
View File

@@ -0,0 +1,67 @@
package com.rj.image;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;
import com.alibaba.dashscope.utils.JsonUtils;
import java.io.IOException;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
public class QwenImage {
static {
// 以下为北京地域url，若使用新加坡地域的模型，需将url替换为：https://dashscope-intl.aliyuncs.com/api/v1
Constants.baseHttpApiUrl = "https://dashscope.aliyuncs.com/api/v1";
}
// 若没有配置环境变量，请用百炼API Key将下行替换为：apiKey="sk-xxx"
// 新加坡和北京地域的API Key不同。获取API Key：https://help.aliyun.com/zh/model-studio/get-api-key
static String apiKey = System.getenv("DASHSCOPE_API_KEY");
public static void call() throws ApiException, NoApiKeyException, UploadFileException, IOException {
MultiModalConversation conv = new MultiModalConversation();
String content1 = "一张在海滩上拍摄的照片。照片中是一位穿着浅色比基尼的女性，她站在海边，背景是海浪、沙滩和远处的山丘。天空晴朗，阳光明媚，整体环境显得非常自然和放松。微笑着，一只手抚摸头发，很妩媚，似乎在享受海边的时光。";
String content = "一张在海滩上拍摄的照片。照片中是一位穿着浅色比基尼的女性，她站在海边，背景是海浪、沙滩和远处的山丘。天空晴朗，阳光明媚，整体环境显得非常自然和放松。这位女性面带微笑，双臂微微张开，似乎在享受海边的时光。";
MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
.content(Arrays.asList(
// Collections.singletonMap("text", "一副典雅庄重的对联悬挂于厅堂之中，房间是个安静古典的中式布置，桌子上放着一些青花瓷，对联上左书“义本生知人机同道善思新”，右书“通云赋智乾坤启数高志远”， 横批“智启通义”，字体飘逸，中间挂在一着一副中国风的画作，内容是岳阳楼。")
Collections.singletonMap("text", content)
)).build();
Map<String, Object> parameters = new HashMap<>();
parameters.put("watermark", true);
parameters.put("prompt_extend", true);
parameters.put("negative_prompt", "");
parameters.put("size", "1328*1328");
MultiModalConversationParam param = MultiModalConversationParam.builder()
.apiKey(apiKey)
.model("qwen-image-plus")
.messages(Collections.singletonList(userMessage))
.parameters(parameters)
.build();
System.out.println("开始调用阿里云多模态对话API，参数 ："+ param);
MultiModalConversationResult result = conv.call(param);
System.out.println(JsonUtils.toJson(result));
}
public static void main(String[] args) {
try {
call();
} catch (ApiException | NoApiKeyException | UploadFileException | IOException e) {
System.out.println(e.getMessage());
}
System.exit(0);
}
}

2
src/test/java/com/rj/service/FaceDetectImageCountTest.java
View File

@@ -231,6 +231,8 @@ public class FaceDetectImageCountTest {

2
src/test/java/com/rj/service/TtsRequestLogShortUrlTest.java
View File

@@ -172,6 +172,8 @@ public class TtsRequestLogShortUrlTest {

2
src/test/java/com/rj/service/VideoSynthesisTempUrlTest.java
View File

@@ -150,6 +150,8 @@ public class VideoSynthesisTempUrlTest {

2
src/test/java/com/rj/service/VideoSynthesisVideoNameTest.java
View File

@@ -125,6 +125,8 @@ public class VideoSynthesisVideoNameTest {